* [PATCH net-next v6 1/3] vmcore: add API to collect hardware dump in second kernel
From: Rahul Lakkireddy @ 2018-04-30 10:13 UTC (permalink / raw)
To: netdev, kexec, linux-fsdevel, linux-kernel
Cc: davem, viro, ebiederm, stephen, akpm, torvalds, ganeshgr,
nirranjan, indranil, Rahul Lakkireddy
In-Reply-To: <cover.1525082669.git.rahul.lakkireddy@chelsio.com>
The sequence of actions done by device drivers to append their device
specific hardware/firmware logs to /proc/vmcore are as follows:
1. During probe (before hardware is initialized), device drivers
register to the vmcore module (via vmcore_add_device_dump()), with
callback function, along with buffer size and log name needed for
firmware/hardware log collection.
2. vmcore module allocates the buffer with requested size. It adds
an Elf note and invokes the device driver's registered callback
function.
3. Device driver collects all hardware/firmware logs into the buffer
and returns control back to vmcore module.
Ensure that the device dump buffer size is always aligned to page size
so that it can be mmaped.
Also, rename alloc_elfnotes_buf() to vmcore_alloc_buf() to make it more
generic and reserve NT_VMCOREDD note type to indicate vmcore device
dump.
Suggested-by: Eric Biederman <ebiederm@xmission.com>.
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Signed-off-by: Ganesh Goudar <ganeshgr@chelsio.com>
---
v6:
- Reworked device dump elf note name to contain vendor identifier.
- Added vmcoredd_header that precedes actual dump in the Elf Note.
- Device dump's name is moved inside vmcoredd_header.
v5:
- Removed enabling CONFIG_PROC_VMCORE_DEVICE_DUMP by default and
updated help message to indicate that the driver must be present
in second kernel's initramfs to collect the underlying device
snapshot.
v4:
- Made __vmcore_add_device_dump() static.
- Moved compile check to define vmcore_add_device_dump() to
crash_dump.h to fix compilation when vmcore.c is not compiled in.
- Convert ---help--- to help in Kconfig as indicated by checkpatch.
- Rebased to tip.
v3:
- Dropped sysfs crashdd module.
- Added CONFIG_PROC_VMCORE_DEVICE_DUMP to allow configuring device
dump support.
- Moved logic related to adding dumps from crashdd to vmcore module.
- Rename all crashdd* to vmcoredd*.
v2:
- Added ABI Documentation for crashdd.
- Directly use octal permission instead of macro.
Changes since rfc v2:
- Moved exporting crashdd from procfs to sysfs. Suggested by
Stephen Hemminger <stephen@networkplumber.org>
- Moved code from fs/proc/crashdd.c to fs/crashdd/ directory.
- Replaced all proc API with sysfs API and updated comments.
- Calling driver callback before creating the binary file under
crashdd sysfs.
- Changed binary dump file permission from S_IRUSR to S_IRUGO.
- Changed module name from CRASH_DRIVER_DUMP to CRASH_DEVICE_DUMP.
rfc v2:
- Collecting logs in 2nd kernel instead of during kernel panic.
Suggested by Eric Biederman <ebiederm@xmission.com>.
- Patch added in this series.
fs/proc/Kconfig | 15 ++++
fs/proc/vmcore.c | 168 ++++++++++++++++++++++++++++++++++++++++++---
include/linux/crash_core.h | 10 +++
include/linux/crash_dump.h | 18 +++++
include/linux/kcore.h | 6 ++
include/uapi/linux/elf.h | 1 +
6 files changed, 209 insertions(+), 9 deletions(-)
diff --git a/fs/proc/Kconfig b/fs/proc/Kconfig
index 1ade1206bb89..0eaeb41453f5 100644
--- a/fs/proc/Kconfig
+++ b/fs/proc/Kconfig
@@ -43,6 +43,21 @@ config PROC_VMCORE
help
Exports the dump image of crashed kernel in ELF format.
+config PROC_VMCORE_DEVICE_DUMP
+ bool "Device Hardware/Firmware Log Collection"
+ depends on PROC_VMCORE
+ default n
+ help
+ After kernel panic, device drivers can collect the device
+ specific snapshot of their hardware or firmware before the
+ underlying devices are initialized in crash recovery kernel.
+ Note that the device driver must be present in the crash
+ recovery kernel's initramfs to collect its underlying device
+ snapshot.
+
+ If you say Y here, the collected device dumps will be added
+ as ELF notes to /proc/vmcore.
+
config PROC_SYSCTL
bool "Sysctl support (/proc/sys)" if EXPERT
depends on PROC_FS
diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c
index a45f0af22a60..2392be007228 100644
--- a/fs/proc/vmcore.c
+++ b/fs/proc/vmcore.c
@@ -20,6 +20,7 @@
#include <linux/init.h>
#include <linux/crash_dump.h>
#include <linux/list.h>
+#include <linux/mutex.h>
#include <linux/vmalloc.h>
#include <linux/pagemap.h>
#include <linux/uaccess.h>
@@ -44,6 +45,12 @@ static u64 vmcore_size;
static struct proc_dir_entry *proc_vmcore;
+#ifdef CONFIG_PROC_VMCORE_DEVICE_DUMP
+/* Device Dump list and mutex to synchronize access to list */
+static LIST_HEAD(vmcoredd_list);
+static DEFINE_MUTEX(vmcoredd_mutex);
+#endif /* CONFIG_PROC_VMCORE_DEVICE_DUMP */
+
/*
* Returns > 0 for RAM pages, 0 for non-RAM pages, < 0 on error
* The called function has to take care of module refcounting.
@@ -302,10 +309,8 @@ static const struct vm_operations_struct vmcore_mmap_ops = {
};
/**
- * alloc_elfnotes_buf - allocate buffer for ELF note segment in
- * vmalloc memory
- *
- * @notes_sz: size of buffer
+ * vmcore_alloc_buf - allocate buffer in vmalloc memory
+ * @sizez: size of buffer
*
* If CONFIG_MMU is defined, use vmalloc_user() to allow users to mmap
* the buffer to user-space by means of remap_vmalloc_range().
@@ -313,12 +318,12 @@ static const struct vm_operations_struct vmcore_mmap_ops = {
* If CONFIG_MMU is not defined, use vzalloc() since mmap_vmcore() is
* disabled and there's no need to allow users to mmap the buffer.
*/
-static inline char *alloc_elfnotes_buf(size_t notes_sz)
+static inline char *vmcore_alloc_buf(size_t size)
{
#ifdef CONFIG_MMU
- return vmalloc_user(notes_sz);
+ return vmalloc_user(size);
#else
- return vzalloc(notes_sz);
+ return vzalloc(size);
#endif
}
@@ -665,7 +670,7 @@ static int __init merge_note_headers_elf64(char *elfptr, size_t *elfsz,
return rc;
*notes_sz = roundup(phdr_sz, PAGE_SIZE);
- *notes_buf = alloc_elfnotes_buf(*notes_sz);
+ *notes_buf = vmcore_alloc_buf(*notes_sz);
if (!*notes_buf)
return -ENOMEM;
@@ -851,7 +856,7 @@ static int __init merge_note_headers_elf32(char *elfptr, size_t *elfsz,
return rc;
*notes_sz = roundup(phdr_sz, PAGE_SIZE);
- *notes_buf = alloc_elfnotes_buf(*notes_sz);
+ *notes_buf = vmcore_alloc_buf(*notes_sz);
if (!*notes_buf)
return -ENOMEM;
@@ -1145,6 +1150,148 @@ static int __init parse_crash_elf_headers(void)
return 0;
}
+#ifdef CONFIG_PROC_VMCORE_DEVICE_DUMP
+/**
+ * vmcoredd_get_note_size - Get size of the note that will be inserted at
+ * beginning of the dump's buffer.
+ * @name: Note's name
+ *
+ * Gets the overall size of the note that will be inserted at the beginning
+ * of the dump's buffer. It also adds padding, if necessary to meet
+ * alignment requirements.
+ */
+static inline size_t vmcoredd_get_note_size(const char *name)
+{
+ /* Must include '\0' and hence the extra +1 */
+ return CRASH_CORE_NOTE_HEAD_BYTES +
+ ALIGN(strlen(name) + 1, sizeof(Elf_Word));
+}
+
+/**
+ * vmcoredd_write_note - Write note at the beginning of the dump's buffer
+ * @buf: Output buffer where the note is written
+ * @data: Dump info
+ * @size: Size of the dump including vmcore device dump header
+ *
+ * Fills beginning of the dump's data with elf note and vmcore device dump
+ * header.
+ */
+static void vmcoredd_write_note(void *buf, struct vmcoredd_data *data,
+ size_t size)
+{
+ struct elf_note *note = (struct elf_note *)buf;
+ struct vmcoredd_header vdd_hdr = { 0 };
+ u8 *note_ptr = buf;
+
+ vdd_hdr.header_len = sizeof(struct vmcoredd_header);
+ vdd_hdr.data_len = data->size;
+ strncpy(vdd_hdr.dump_name, data->dump_name, sizeof(vdd_hdr.dump_name));
+
+ note->n_namesz = vmcoredd_get_note_size(data->note_name) -
+ CRASH_CORE_NOTE_HEAD_BYTES;
+ note->n_descsz = size;
+ note->n_type = NT_VMCOREDD;
+
+ /* Write note name */
+ note_ptr += CRASH_CORE_NOTE_HEAD_BYTES;
+ strncpy((char *)note_ptr, data->note_name, note->n_namesz);
+
+ /* Write vmcore device dump header */
+ note_ptr += note->n_namesz;
+ memcpy(note_ptr, &vdd_hdr, sizeof(vdd_hdr));
+}
+
+/**
+ * vmcore_add_device_dump - Add a buffer containing device dump to vmcore
+ * @data: dump info.
+ *
+ * Allocate a buffer and invoke the calling driver's dump collect routine.
+ * Write Elf note at the beginning of the buffer to indicate vmcore device
+ * dump and add the dump to global list.
+ */
+static int __vmcore_add_device_dump(struct vmcoredd_data *data)
+{
+ size_t note_size, data_size;
+ struct vmcoredd_node *dump;
+ void *buf = NULL;
+ int ret;
+
+ if (!data || !strlen(data->note_name) ||
+ !strlen(data->dump_name) ||
+ !data->vmcoredd_callback || !data->size)
+ return -EINVAL;
+
+ dump = vzalloc(sizeof(*dump));
+ if (!dump) {
+ ret = -ENOMEM;
+ goto out_err;
+ }
+
+ note_size = vmcoredd_get_note_size(data->note_name);
+ /* Keep size of the buffer page aligned so that it can be mmaped */
+ data_size = roundup(note_size + sizeof(struct vmcoredd_header) +
+ data->size, PAGE_SIZE);
+
+ /* Allocate buffer for driver's to write their dumps */
+ buf = vmcore_alloc_buf(data_size);
+ if (!buf) {
+ ret = -ENOMEM;
+ goto out_err;
+ }
+
+ vmcoredd_write_note(buf, data, data_size - note_size);
+
+ /* Invoke the driver's dump collection routing */
+ ret = data->vmcoredd_callback(data, buf + note_size +
+ sizeof(struct vmcoredd_header));
+ if (ret)
+ goto out_err;
+
+ dump->buf = buf;
+ dump->size = data_size;
+
+ /* Add the dump to driver sysfs list */
+ mutex_lock(&vmcoredd_mutex);
+ list_add_tail(&dump->list, &vmcoredd_list);
+ mutex_unlock(&vmcoredd_mutex);
+
+ return 0;
+
+out_err:
+ if (buf)
+ vfree(buf);
+
+ if (dump)
+ vfree(dump);
+
+ return ret;
+}
+
+int vmcore_add_device_dump(struct vmcoredd_data *data)
+{
+ return __vmcore_add_device_dump(data);
+}
+EXPORT_SYMBOL(vmcore_add_device_dump);
+#endif /* CONFIG_PROC_VMCORE_DEVICE_DUMP */
+
+/* Free all dumps in vmcore device dump list */
+static void vmcore_free_device_dumps(void)
+{
+#ifdef CONFIG_PROC_VMCORE_DEVICE_DUMP
+ mutex_lock(&vmcoredd_mutex);
+ while (!list_empty(&vmcoredd_list)) {
+ struct vmcoredd_node *dump;
+
+ dump = list_first_entry(&vmcoredd_list, struct vmcoredd_node,
+ list);
+ list_del(&dump->list);
+ vfree(dump->buf);
+ vfree(dump);
+ }
+ mutex_unlock(&vmcoredd_mutex);
+#endif /* CONFIG_PROC_VMCORE_DEVICE_DUMP */
+}
+
/* Init function for vmcore module. */
static int __init vmcore_init(void)
{
@@ -1192,4 +1339,7 @@ void vmcore_cleanup(void)
kfree(m);
}
free_elfcorebuf();
+
+ /* clear vmcore device dump list */
+ vmcore_free_device_dumps();
}
diff --git a/include/linux/crash_core.h b/include/linux/crash_core.h
index b511f6d24b42..e38c5dbe3016 100644
--- a/include/linux/crash_core.h
+++ b/include/linux/crash_core.h
@@ -27,6 +27,16 @@
VMCOREINFO_NOTE_NAME_BYTES + \
VMCOREINFO_BYTES)
+#define VMCOREDD_MAX_NAME_BYTES 32
+
+struct vmcoredd_header {
+ unsigned int header_len; /* Length of this header */
+ unsigned int reserved;
+ unsigned long data_len; /* Length of the device dump */
+ char dump_name[VMCOREDD_MAX_NAME_BYTES]; /* Name of the device dump */
+ unsigned char reserved2[16];
+};
+
typedef u32 note_buf_t[CRASH_CORE_NOTE_BYTES/4];
void crash_update_vmcoreinfo_safecopy(void *ptr);
diff --git a/include/linux/crash_dump.h b/include/linux/crash_dump.h
index f7ac2aa93269..45ce7ca4625a 100644
--- a/include/linux/crash_dump.h
+++ b/include/linux/crash_dump.h
@@ -93,4 +93,22 @@ static inline bool is_kdump_kernel(void) { return 0; }
#endif /* CONFIG_CRASH_DUMP */
extern unsigned long saved_max_pfn;
+
+/* Device Dump information to be filled by drivers */
+struct vmcoredd_data {
+ char note_name[VMCOREDD_MAX_NAME_BYTES]; /* Name of the Elf note */
+ char dump_name[VMCOREDD_MAX_NAME_BYTES]; /* Unique name of the dump */
+ unsigned long size; /* Size of the dump */
+ /* Driver's registered callback to be invoked to collect dump */
+ int (*vmcoredd_callback)(struct vmcoredd_data *data, void *buf);
+};
+
+#ifdef CONFIG_PROC_VMCORE_DEVICE_DUMP
+int vmcore_add_device_dump(struct vmcoredd_data *data);
+#else
+static inline int vmcore_add_device_dump(struct vmcoredd_data *data)
+{
+ return -EOPNOTSUPP;
+}
+#endif /* CONFIG_PROC_VMCORE_DEVICE_DUMP */
#endif /* LINUX_CRASHDUMP_H */
diff --git a/include/linux/kcore.h b/include/linux/kcore.h
index 80db19d3a505..aa26e7199060 100644
--- a/include/linux/kcore.h
+++ b/include/linux/kcore.h
@@ -28,6 +28,12 @@ struct vmcore {
loff_t offset;
};
+struct vmcoredd_node {
+ struct list_head list; /* List of dumps */
+ void *buf; /* Buffer containing device's dump */
+ unsigned long size; /* Size of the buffer */
+};
+
#ifdef CONFIG_PROC_KCORE
extern void kclist_add(struct kcore_list *, void *, size_t, int type);
#else
diff --git a/include/uapi/linux/elf.h b/include/uapi/linux/elf.h
index e2535d6dcec7..4e12c423b9fe 100644
--- a/include/uapi/linux/elf.h
+++ b/include/uapi/linux/elf.h
@@ -421,6 +421,7 @@ typedef struct elf64_shdr {
#define NT_ARM_SYSTEM_CALL 0x404 /* ARM system call number */
#define NT_ARM_SVE 0x405 /* ARM Scalable Vector Extension registers */
#define NT_ARC_V2 0x600 /* ARCv2 accumulator/extra registers */
+#define NT_VMCOREDD 0x700 /* Vmcore Device Dump Note */
/* Note header in a PT_NOTE section */
typedef struct elf32_note {
--
2.14.1
^ permalink raw reply related
* [PATCH net-next v6 0/3] kernel: add support to collect hardware logs in crash recovery kernel
From: Rahul Lakkireddy @ 2018-04-30 10:13 UTC (permalink / raw)
To: netdev-u79uwXL29TY76Z2rM5mHXA,
kexec-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r,
linux-fsdevel-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
Cc: indranil-ut6Up61K2wZBDgjK7y7TUQ, nirranjan-ut6Up61K2wZBDgjK7y7TUQ,
stephen-OTpzqLSitTUnbdJkjeBofR2eb7JE58TQ,
ganeshgr-ut6Up61K2wZBDgjK7y7TUQ, Rahul Lakkireddy,
ebiederm-aS9lmoZGLiVWk0Htik3J/w,
akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b,
torvalds-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b,
davem-fT/PcQaiUtIeIZ0/mPfg9Q,
viro-RmSDqhL/yNMiFSDQTTA3OLVCufUGDwFn
On production servers running variety of workloads over time, kernel
panic can happen sporadically after days or even months. It is
important to collect as much debug logs as possible to root cause
and fix the problem, that may not be easy to reproduce. Snapshot of
underlying hardware/firmware state (like register dump, firmware
logs, adapter memory, etc.), at the time of kernel panic will be very
helpful while debugging the culprit device driver.
This series of patches add new generic framework that enable device
drivers to collect device specific snapshot of the hardware/firmware
state of the underlying device in the crash recovery kernel. In crash
recovery kernel, the collected logs are added as elf notes to
/proc/vmcore, which is copied by user space scripts for post-analysis.
The sequence of actions done by device drivers to append their device
specific hardware/firmware logs to /proc/vmcore are as follows:
1. During probe (before hardware is initialized), device drivers
register to the vmcore module (via vmcore_add_device_dump()), with
callback function, along with buffer size and log name needed for
firmware/hardware log collection.
2. vmcore module allocates the buffer with requested size. It adds
an elf note and invokes the device driver's registered callback
function.
3. Device driver collects all hardware/firmware logs into the buffer
and returns control back to vmcore module.
The device specific hardware/firmware logs can be seen as elf notes:
# readelf -n /proc/vmcore
Displaying notes found at file offset 0x00001000 with length 0x040032bc:
Owner Data size Description
CHELSIO 0x02000fec Unknown note type: (0x00000700)
CHELSIO 0x02000fec Unknown note type: (0x00000700)
CORE 0x00000150 NT_PRSTATUS (prstatus structure)
CORE 0x00000150 NT_PRSTATUS (prstatus structure)
CORE 0x00000150 NT_PRSTATUS (prstatus structure)
CORE 0x00000150 NT_PRSTATUS (prstatus structure)
CORE 0x00000150 NT_PRSTATUS (prstatus structure)
CORE 0x00000150 NT_PRSTATUS (prstatus structure)
CORE 0x00000150 NT_PRSTATUS (prstatus structure)
CORE 0x00000150 NT_PRSTATUS (prstatus structure)
VMCOREINFO 0x00000784 Unknown note type: (0x00000000)
Patch 1 adds API to vmcore module to allow drivers to register callback
to collect the device specific hardware/firmware logs. The logs will
be added to /proc/vmcore as elf notes.
Patch 2 updates read and mmap logic to append device specific hardware/
firmware logs as elf notes.
Patch 3 shows a cxgb4 driver example using the API to collect
hardware/firmware logs in crash recovery kernel, before hardware is
initialized.
Thanks,
Rahul
---
v6:
- Reworked device dump elf note name to contain vendor identifier.
- Added vmcoredd_header that precedes actual dump in the Elf Note.
- Device dump's name is moved inside vmcoredd_header.
- Added "CHELSIO" string as vendor identifier in the Elf Note name
for cxgb4 device dumps.
v5:
- Removed enabling CONFIG_PROC_VMCORE_DEVICE_DUMP by default and
updated help message.
v4:
- Made __vmcore_add_device_dump() static.
- Moved compile check to define vmcore_add_device_dump() to
crash_dump.h to fix compilation when vmcore.c is not compiled in.
- Convert ---help--- to help in Kconfig as indicated by checkpatch.
- Rebased to tip.
v3:
- Dropped sysfs crashdd module.
- Exported dumps as elf notes. Suggested by Eric Biederman
<ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>. Added as patch 2 in this version.
- Added CONFIG_PROC_VMCORE_DEVICE_DUMP to allow configuring device
dump support.
- Moved logic related to adding dumps from crashdd to vmcore module.
- Rename all crashdd* to vmcoredd*.
- Updated comments.
v2:
- Added ABI Documentation for crashdd.
- Directly use octal permission instead of macro.
Changes since rfc v2:
- Moved exporting crashdd from procfs to sysfs. Suggested by
Stephen Hemminger <stephen-OTpzqLSitTUnbdJkjeBofR2eb7JE58TQ@public.gmane.org>
- Moved code from fs/proc/crashdd.c to fs/crashdd/ directory.
- Replaced all proc API with sysfs API and updated comments.
- Calling driver callback before creating the binary file under
crashdd sysfs.
- Changed binary dump file permission from S_IRUSR to S_IRUGO.
- Changed module name from CRASH_DRIVER_DUMP to CRASH_DEVICE_DUMP.
rfc v2:
- Collecting logs in 2nd kernel instead of during kernel panic.
Suggested by Eric Biederman <ebiederm-aS9lmoZGLiVWk0Htik3J/w@public.gmane.org>.
- Added new crashdd module that exports /proc/crashdd/ containing
driver's registered hardware/firmware logs in patch 1.
- Replaced the API to allow drivers to register their hardware/firmware
log collect routine in crash recovery kernel in patch 1.
- Updated patch 2 to use the new API in patch 1.
Rahul Lakkireddy (3):
vmcore: add API to collect hardware dump in second kernel
vmcore: append device dumps to vmcore as elf notes
cxgb4: collect hardware dump in second kernel
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 4 +
drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.c | 27 ++
drivers/net/ethernet/chelsio/cxgb4/cxgb4_cudbg.h | 4 +
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 10 +
fs/proc/Kconfig | 15 +
fs/proc/vmcore.c | 415 ++++++++++++++++++++++-
include/linux/crash_core.h | 10 +
include/linux/crash_dump.h | 18 +
include/linux/kcore.h | 6 +
include/uapi/linux/elf.h | 1 +
10 files changed, 497 insertions(+), 13 deletions(-)
--
2.14.1
^ permalink raw reply
* Re: [PATCH v2 1/2] selftests/bpf: Makefile: add includes to fix broken test build
From: Daniel Borkmann @ 2018-04-30 10:12 UTC (permalink / raw)
To: Sirio Balmelli; +Cc: netdev
In-Reply-To: <20180430071730.nghfksp2tkj3x5ld@vm4>
On 04/30/2018 09:17 AM, Sirio Balmelli wrote:
> Build fails with missing 'asm/bitsperlong.h'.
>
> Include this from tools/arch/[arch]/include/uapi
> using an architecture-specific include sourced from 'Module.symvers'
> (is this legitimate? another idea was to do a `uname -p` but what if
> caller was cross-compiling?)
To point to an arch specific include, the way this works under tools
can be seen e.g. in barrier.h, see tools/include/asm/barrier.h. This
will then pull in one of the tools/arch/*/include/asm/barrier.h. We
would need to do something similar for tools/arch/*/include/uapi/asm/bitsperlong.h
Then you also don't need the Module.symvers hack to derive the underlying
arch. See 720f228e8d31 ("bpf: fix broken BPF selftest build") for another,
similar example we had in the past.
> Once the arch-specific include is fixed, a lack of 'asm/byteorder.h'
> now rears its ugly head.
> The only sane (ie: will not pull in kernel headers) place I can find it
> is in $(ROOT)/usr/include - add this to the include.
Why it doesn't work adding a new file tools/include/uapi/asm/byteorder.h
that points to one present in tools/arch/*/include/uapi/asm/bitsperlong.h
or tools/include/uapi/asm-generic/bitsperlong.h otherwise?
> Note that $(ROOT)/usr/include is generated on kernel build, hence my
> assumption that 'Module.symvers' will also be present.
The problem with this approach is that you need to do 'make header_install'
in order to resolve this dependency, whereas tools infrastructure has no
such dependency. Here's some more background context on the tools infra
that perhaps helps why it's done this way:
https://www.spinics.net/lists/netdev/msg487309.html
Please see to add a tools/include/uapi/asm/byteorder.h to fix the issue.
Thanks a lot,
Daniel
^ permalink raw reply
* Re: [PATCH net] ipv6: fix uninit-value in ip6_multipath_l3_keys()
From: Jakub Sitnicki @ 2018-04-30 9:59 UTC (permalink / raw)
To: Eric Dumazet; +Cc: David S . Miller, netdev, Eric Dumazet
In-Reply-To: <20180429165459.164538-1-edumazet@google.com>
On Sun, 29 Apr 2018 09:54:59 -0700
Eric Dumazet <edumazet@google.com> wrote:
> syzbot/KMSAN reported an uninit-value in ip6_multipath_l3_keys(),
> root caused to a bad assumption of ICMP header being already
> pulled in skb->head
>
> ip_multipath_l3_keys() does the correct thing, so it is an IPv6 only bug.
>
> BUG: KMSAN: uninit-value in ip6_multipath_l3_keys net/ipv6/route.c:1830 [inline]
> BUG: KMSAN: uninit-value in rt6_multipath_hash+0x5c4/0x640 net/ipv6/route.c:1858
> CPU: 0 PID: 4507 Comm: syz-executor661 Not tainted 4.16.0+ #87
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> Call Trace:
> __dump_stack lib/dump_stack.c:17 [inline]
> dump_stack+0x185/0x1d0 lib/dump_stack.c:53
> kmsan_report+0x142/0x240 mm/kmsan/kmsan.c:1067
> __msan_warning_32+0x6c/0xb0 mm/kmsan/kmsan_instr.c:683
> ip6_multipath_l3_keys net/ipv6/route.c:1830 [inline]
> rt6_multipath_hash+0x5c4/0x640 net/ipv6/route.c:1858
> ip6_route_input+0x65a/0x920 net/ipv6/route.c:1884
> ip6_rcv_finish+0x413/0x6e0 net/ipv6/ip6_input.c:69
> NF_HOOK include/linux/netfilter.h:288 [inline]
> ipv6_rcv+0x1e16/0x2340 net/ipv6/ip6_input.c:208
> __netif_receive_skb_core+0x47df/0x4a90 net/core/dev.c:4562
> __netif_receive_skb net/core/dev.c:4627 [inline]
> netif_receive_skb_internal+0x49d/0x630 net/core/dev.c:4701
> netif_receive_skb+0x230/0x240 net/core/dev.c:4725
> tun_rx_batched drivers/net/tun.c:1555 [inline]
> tun_get_user+0x740f/0x7c60 drivers/net/tun.c:1962
> tun_chr_write_iter+0x1d4/0x330 drivers/net/tun.c:1990
> call_write_iter include/linux/fs.h:1782 [inline]
> new_sync_write fs/read_write.c:469 [inline]
> __vfs_write+0x7fb/0x9f0 fs/read_write.c:482
> vfs_write+0x463/0x8d0 fs/read_write.c:544
> SYSC_write+0x172/0x360 fs/read_write.c:589
> SyS_write+0x55/0x80 fs/read_write.c:581
> do_syscall_64+0x309/0x430 arch/x86/entry/common.c:287
> entry_SYSCALL_64_after_hwframe+0x3d/0xa2
>
> Fixes: 23aebdacb05d ("ipv6: Compute multipath hash for ICMP errors from offending packet")
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Reported-by: syzbot <syzkaller@googlegroups.com>
> Cc: Jakub Sitnicki <jkbs@redhat.com>
> ---
> net/ipv6/route.c | 7 ++++++-
> 1 file changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/net/ipv6/route.c b/net/ipv6/route.c
> index cde7d8251377c1a115e02c46843d361d3c0b4313..f4d61736c41abe8cd7f439c4a37100e90c1eacca 100644
> --- a/net/ipv6/route.c
> +++ b/net/ipv6/route.c
> @@ -1835,11 +1835,16 @@ static void ip6_multipath_l3_keys(const struct sk_buff *skb,
> const struct ipv6hdr *inner_iph;
> const struct icmp6hdr *icmph;
> struct ipv6hdr _inner_iph;
> + struct icmp6hdr _icmph;
>
> if (likely(outer_iph->nexthdr != IPPROTO_ICMPV6))
> goto out;
>
> - icmph = icmp6_hdr(skb);
> + icmph = skb_header_pointer(skb, skb_transport_offset(skb),
> + sizeof(_icmph), &_icmph);
> + if (!icmph)
> + goto out;
> +
> if (icmph->icmp6_type != ICMPV6_DEST_UNREACH &&
> icmph->icmp6_type != ICMPV6_PKT_TOOBIG &&
> icmph->icmp6_type != ICMPV6_TIME_EXCEED &&
Thank you for the fix. I was not being careful there.
Acked-by: Jakub Sitnicki <jkbs@redhat.com>
^ permalink raw reply
* Re: [PATCH net-next] cls_flower: Support multiple masks per priority
From: Jiri Pirko @ 2018-04-30 9:44 UTC (permalink / raw)
To: Paul Blakey
Cc: Jiri Pirko, Cong Wang, Jamal Hadi Salim, David Miller, netdev,
Yevgeny Kliteynik, Roi Dayan, Shahar Klein, Mark Bloch,
Or Gerlitz
In-Reply-To: <1525076909-8767-1-git-send-email-paulb@mellanox.com>
Mon, Apr 30, 2018 at 10:28:29AM CEST, paulb@mellanox.com wrote:
>Currently flower doesn't support inserting filters with different masks
>on a single priority, even if the actual flows (key + mask) inserted
>aren't overlapping, as with the use case of offloading openvswitch
>datapath flows. Instead one must go up one level, and assign different
>priorities for each mask, which will create a different flower
>instances.
>
>This patch opens flower to support more than one mask per priority,
>and a single flower instance. It does so by adding another hash table
>on top of the existing one which will store the different masks,
>and the filters that share it.
>
>The user is left with the responsibilty of ensuring non overlapping
s/responsibilty/responsibility
>flows, otherwise precedence is not guaranteed.
>
>Signed-off-by: Paul Blakey <paulb@mellanox.com>
Looks good to me.
One small nit below.
Feel free to append my tag to v2:
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
[...]
>@@ -103,15 +113,22 @@ static void fl_mask_update_range(struct fl_flow_mask *mask)
> {
> const u8 *bytes = (const u8 *) &mask->key;
> size_t size = sizeof(mask->key);
>- size_t i, first = 0, last = size - 1;
>+ size_t i, first = 0, last;
>
>- for (i = 0; i < sizeof(mask->key); i++) {
>+ for (i = 0; i < size; i++) {
>+ if (bytes[i]) {
>+ first = i;
>+ break;
>+ }
>+ }
>+ last = first;
>+ for (i = size - 1; i != first; i--) {
> if (bytes[i]) {
>- if (!first && i)
>- first = i;
> last = i;
>+ break;
> }
> }
>+
Remove this newline.
> mask->range.start = rounddown(first, sizeof(long));
> mask->range.end = roundup(last + 1, sizeof(long));
> }
[...]
^ permalink raw reply
* Re: [PATCH] net: gemini: fix gmac_start_xmit()'s return type
From: Linus Walleij @ 2018-04-30 9:17 UTC (permalink / raw)
To: Luc Van Oostenryck
Cc: linux-kernel@vger.kernel.org, Hans Ulli Kroll, Linux ARM, netdev
In-Reply-To: <20180424131642.3806-1-luc.vanoostenryck@gmail.com>
On Tue, Apr 24, 2018 at 3:16 PM, Luc Van Oostenryck
<luc.vanoostenryck@gmail.com> wrote:
> The method ndo_start_xmit() is defined as returning an 'netdev_tx_t',
> which is a typedef for an enum type, but the implementation in this
> driver returns an 'int'.
>
> Fix this by returning 'netdev_tx_t' in this driver too.
>
> Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Thanks for fixing this!
Linus Walleij
^ permalink raw reply
* Re: [PATCH bpf-next] selftests/bpf: bpf tunnel test.
From: Daniel Borkmann @ 2018-04-30 9:05 UTC (permalink / raw)
To: Y Song, William Tu; +Cc: Linux Kernel Network Developers
In-Reply-To: <CAH3MdRVL6=5N1yKf2w7tAiGCV+48XMMtQ9iq4VuSC_-Z7YL=nQ@mail.gmail.com>
On 04/30/2018 09:02 AM, Y Song wrote:
> Hi, William,
>
> When compiled the selftests/bpf in my centos 7 based system, I have
> the following failures,
>
> clang -I. -I./include/uapi -I../../../include/uapi
> -Wno-compare-distinct-pointer-types \
> -O2 -target bpf -emit-llvm -c test_tunnel_kern.c -o - | \
> llc -march=bpf -mcpu=generic -filetype=obj -o
> /data/users/yhs/work/net-next/tools/testing/selftests/bpf/test_tunnel_kern.o
> test_tunnel_kern.c:21:10: fatal error: 'linux/erspan.h' file not found
> #include <linux/erspan.h>
> ^~~~~~~~~~~~~~~~
> 1 error generated.
>
> Maybe I missed some packages to install?
Looks like we need a copy for tools infrastructure in tools/include/uapi/linux/erspan.h.
William, could you take a look?
Thanks,
Daniel
^ permalink raw reply
* [PATCH net-next] cls_flower: Support multiple masks per priority
From: Paul Blakey @ 2018-04-30 8:28 UTC (permalink / raw)
To: Jiri Pirko, Cong Wang, Jamal Hadi Salim, David Miller, netdev
Cc: Yevgeny Kliteynik, Roi Dayan, Shahar Klein, Mark Bloch,
Or Gerlitz, Paul Blakey
Currently flower doesn't support inserting filters with different masks
on a single priority, even if the actual flows (key + mask) inserted
aren't overlapping, as with the use case of offloading openvswitch
datapath flows. Instead one must go up one level, and assign different
priorities for each mask, which will create a different flower
instances.
This patch opens flower to support more than one mask per priority,
and a single flower instance. It does so by adding another hash table
on top of the existing one which will store the different masks,
and the filters that share it.
The user is left with the responsibilty of ensuring non overlapping
flows, otherwise precedence is not guaranteed.
Signed-off-by: Paul Blakey <paulb@mellanox.com>
---
net/sched/cls_flower.c | 279 +++++++++++++++++++++++++++++++------------------
1 file changed, 179 insertions(+), 100 deletions(-)
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index d964e60..258b385 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -61,16 +61,18 @@ struct fl_flow_mask_range {
struct fl_flow_mask {
struct fl_flow_key key;
struct fl_flow_mask_range range;
- struct rcu_head rcu;
+ struct rhash_head ht_node;
+ struct rhashtable ht;
+ struct rhashtable_params filter_ht_params;
+ struct flow_dissector dissector;
+ struct list_head filters;
+ struct rcu_head rcu;
+ struct list_head list;
};
struct cls_fl_head {
struct rhashtable ht;
- struct fl_flow_mask mask;
- struct flow_dissector dissector;
- bool mask_assigned;
- struct list_head filters;
- struct rhashtable_params ht_params;
+ struct list_head masks;
union {
struct work_struct work;
struct rcu_head rcu;
@@ -79,6 +81,7 @@ struct cls_fl_head {
};
struct cls_fl_filter {
+ struct fl_flow_mask *mask;
struct rhash_head ht_node;
struct fl_flow_key mkey;
struct tcf_exts exts;
@@ -94,6 +97,13 @@ struct cls_fl_filter {
struct net_device *hw_dev;
};
+static const struct rhashtable_params mask_ht_params = {
+ .key_offset = offsetof(struct fl_flow_mask, key),
+ .key_len = sizeof(struct fl_flow_key),
+ .head_offset = offsetof(struct fl_flow_mask, ht_node),
+ .automatic_shrinking = true,
+};
+
static unsigned short int fl_mask_range(const struct fl_flow_mask *mask)
{
return mask->range.end - mask->range.start;
@@ -103,15 +113,22 @@ static void fl_mask_update_range(struct fl_flow_mask *mask)
{
const u8 *bytes = (const u8 *) &mask->key;
size_t size = sizeof(mask->key);
- size_t i, first = 0, last = size - 1;
+ size_t i, first = 0, last;
- for (i = 0; i < sizeof(mask->key); i++) {
+ for (i = 0; i < size; i++) {
+ if (bytes[i]) {
+ first = i;
+ break;
+ }
+ }
+ last = first;
+ for (i = size - 1; i != first; i--) {
if (bytes[i]) {
- if (!first && i)
- first = i;
last = i;
+ break;
}
}
+
mask->range.start = rounddown(first, sizeof(long));
mask->range.end = roundup(last + 1, sizeof(long));
}
@@ -140,12 +157,11 @@ static void fl_clear_masked_range(struct fl_flow_key *key,
memset(fl_key_get_start(key, mask), 0, fl_mask_range(mask));
}
-static struct cls_fl_filter *fl_lookup(struct cls_fl_head *head,
+static struct cls_fl_filter *fl_lookup(struct fl_flow_mask *mask,
struct fl_flow_key *mkey)
{
- return rhashtable_lookup_fast(&head->ht,
- fl_key_get_start(mkey, &head->mask),
- head->ht_params);
+ return rhashtable_lookup_fast(&mask->ht, fl_key_get_start(mkey, mask),
+ mask->filter_ht_params);
}
static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp,
@@ -153,28 +169,28 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp,
{
struct cls_fl_head *head = rcu_dereference_bh(tp->root);
struct cls_fl_filter *f;
+ struct fl_flow_mask *mask;
struct fl_flow_key skb_key;
struct fl_flow_key skb_mkey;
- if (!atomic_read(&head->ht.nelems))
- return -1;
-
- fl_clear_masked_range(&skb_key, &head->mask);
+ list_for_each_entry_rcu(mask, &head->masks, list) {
+ fl_clear_masked_range(&skb_key, mask);
- skb_key.indev_ifindex = skb->skb_iif;
- /* skb_flow_dissect() does not set n_proto in case an unknown protocol,
- * so do it rather here.
- */
- skb_key.basic.n_proto = skb->protocol;
- skb_flow_dissect_tunnel_info(skb, &head->dissector, &skb_key);
- skb_flow_dissect(skb, &head->dissector, &skb_key, 0);
+ skb_key.indev_ifindex = skb->skb_iif;
+ /* skb_flow_dissect() does not set n_proto in case an unknown
+ * protocol, so do it rather here.
+ */
+ skb_key.basic.n_proto = skb->protocol;
+ skb_flow_dissect_tunnel_info(skb, &mask->dissector, &skb_key);
+ skb_flow_dissect(skb, &mask->dissector, &skb_key, 0);
- fl_set_masked_key(&skb_mkey, &skb_key, &head->mask);
+ fl_set_masked_key(&skb_mkey, &skb_key, mask);
- f = fl_lookup(head, &skb_mkey);
- if (f && !tc_skip_sw(f->flags)) {
- *res = f->res;
- return tcf_exts_exec(skb, &f->exts, res);
+ f = fl_lookup(mask, &skb_mkey);
+ if (f && !tc_skip_sw(f->flags)) {
+ *res = f->res;
+ return tcf_exts_exec(skb, &f->exts, res);
+ }
}
return -1;
}
@@ -182,18 +198,40 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp,
static int fl_init(struct tcf_proto *tp)
{
struct cls_fl_head *head;
+ int err;
head = kzalloc(sizeof(*head), GFP_KERNEL);
if (!head)
return -ENOBUFS;
- INIT_LIST_HEAD_RCU(&head->filters);
+ INIT_LIST_HEAD_RCU(&head->masks);
rcu_assign_pointer(tp->root, head);
idr_init(&head->handle_idr);
+ err = rhashtable_init(&head->ht, &mask_ht_params);
+ if (err)
+ return err;
+
return 0;
}
+static bool fl_mask_put(struct cls_fl_head *head, struct fl_flow_mask *mask,
+ bool async)
+{
+ if (!list_empty(&mask->filters))
+ return false;
+
+ rhashtable_remove_fast(&head->ht, &mask->ht_node, mask_ht_params);
+ rhashtable_destroy(&mask->ht);
+ list_del_rcu(&mask->list);
+ if (async)
+ kfree_rcu(mask, rcu);
+ else
+ kfree(mask);
+
+ return true;
+}
+
static void __fl_destroy_filter(struct cls_fl_filter *f)
{
tcf_exts_destroy(&f->exts);
@@ -234,8 +272,6 @@ static void fl_hw_destroy_filter(struct tcf_proto *tp, struct cls_fl_filter *f,
}
static int fl_hw_replace_filter(struct tcf_proto *tp,
- struct flow_dissector *dissector,
- struct fl_flow_key *mask,
struct cls_fl_filter *f,
struct netlink_ext_ack *extack)
{
@@ -247,8 +283,8 @@ static int fl_hw_replace_filter(struct tcf_proto *tp,
tc_cls_common_offload_init(&cls_flower.common, tp, f->flags, extack);
cls_flower.command = TC_CLSFLOWER_REPLACE;
cls_flower.cookie = (unsigned long) f;
- cls_flower.dissector = dissector;
- cls_flower.mask = mask;
+ cls_flower.dissector = &f->mask->dissector;
+ cls_flower.mask = &f->mask->key;
cls_flower.key = &f->mkey;
cls_flower.exts = &f->exts;
cls_flower.classid = f->res.classid;
@@ -283,28 +319,31 @@ static void fl_hw_update_stats(struct tcf_proto *tp, struct cls_fl_filter *f)
&cls_flower, false);
}
-static void __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f,
+static bool __fl_delete(struct tcf_proto *tp, struct cls_fl_filter *f,
struct netlink_ext_ack *extack)
{
struct cls_fl_head *head = rtnl_dereference(tp->root);
+ bool async = tcf_exts_get_net(&f->exts);
+ bool last;
idr_remove(&head->handle_idr, f->handle);
list_del_rcu(&f->list);
+ last = fl_mask_put(head, f->mask, async);
if (!tc_skip_hw(f->flags))
fl_hw_destroy_filter(tp, f, extack);
tcf_unbind_filter(tp, &f->res);
- if (tcf_exts_get_net(&f->exts))
+ if (async)
call_rcu(&f->rcu, fl_destroy_filter);
else
__fl_destroy_filter(f);
+
+ return last;
}
static void fl_destroy_sleepable(struct work_struct *work)
{
struct cls_fl_head *head = container_of(work, struct cls_fl_head,
work);
- if (head->mask_assigned)
- rhashtable_destroy(&head->ht);
kfree(head);
module_put(THIS_MODULE);
}
@@ -320,10 +359,15 @@ static void fl_destroy_rcu(struct rcu_head *rcu)
static void fl_destroy(struct tcf_proto *tp, struct netlink_ext_ack *extack)
{
struct cls_fl_head *head = rtnl_dereference(tp->root);
+ struct fl_flow_mask *mask, *next_mask;
struct cls_fl_filter *f, *next;
- list_for_each_entry_safe(f, next, &head->filters, list)
- __fl_delete(tp, f, extack);
+ list_for_each_entry_safe(mask, next_mask, &head->masks, list) {
+ list_for_each_entry_safe(f, next, &mask->filters, list) {
+ if (__fl_delete(tp, f, extack))
+ break;
+ }
+ }
idr_destroy(&head->handle_idr);
__module_get(THIS_MODULE);
@@ -715,14 +759,14 @@ static int fl_set_key(struct net *net, struct nlattr **tb,
return ret;
}
-static bool fl_mask_eq(struct fl_flow_mask *mask1,
- struct fl_flow_mask *mask2)
+static void fl_mask_copy(struct fl_flow_mask *dst,
+ struct fl_flow_mask *src)
{
- const long *lmask1 = fl_key_get_start(&mask1->key, mask1);
- const long *lmask2 = fl_key_get_start(&mask2->key, mask2);
+ const void *psrc = fl_key_get_start(&src->key, src);
+ void *pdst = fl_key_get_start(&dst->key, src);
- return !memcmp(&mask1->range, &mask2->range, sizeof(mask1->range)) &&
- !memcmp(lmask1, lmask2, fl_mask_range(mask1));
+ memcpy(pdst, psrc, fl_mask_range(src));
+ dst->range = src->range;
}
static const struct rhashtable_params fl_ht_params = {
@@ -731,14 +775,13 @@ static const struct rhashtable_params fl_ht_params = {
.automatic_shrinking = true,
};
-static int fl_init_hashtable(struct cls_fl_head *head,
- struct fl_flow_mask *mask)
+static int fl_init_mask_hashtable(struct fl_flow_mask *mask)
{
- head->ht_params = fl_ht_params;
- head->ht_params.key_len = fl_mask_range(mask);
- head->ht_params.key_offset += mask->range.start;
+ mask->filter_ht_params = fl_ht_params;
+ mask->filter_ht_params.key_len = fl_mask_range(mask);
+ mask->filter_ht_params.key_offset += mask->range.start;
- return rhashtable_init(&head->ht, &head->ht_params);
+ return rhashtable_init(&mask->ht, &mask->filter_ht_params);
}
#define FL_KEY_MEMBER_OFFSET(member) offsetof(struct fl_flow_key, member)
@@ -761,8 +804,7 @@ static int fl_init_hashtable(struct cls_fl_head *head,
FL_KEY_SET(keys, cnt, id, member); \
} while(0);
-static void fl_init_dissector(struct cls_fl_head *head,
- struct fl_flow_mask *mask)
+static void fl_init_dissector(struct fl_flow_mask *mask)
{
struct flow_dissector_key keys[FLOW_DISSECTOR_KEY_MAX];
size_t cnt = 0;
@@ -802,31 +844,66 @@ static void fl_init_dissector(struct cls_fl_head *head,
FL_KEY_SET_IF_MASKED(&mask->key, keys, cnt,
FLOW_DISSECTOR_KEY_ENC_PORTS, enc_tp);
- skb_flow_dissector_init(&head->dissector, keys, cnt);
+ skb_flow_dissector_init(&mask->dissector, keys, cnt);
+}
+
+static struct fl_flow_mask *fl_create_new_mask(struct cls_fl_head *head,
+ struct fl_flow_mask *mask)
+{
+ struct fl_flow_mask *newmask;
+ int err;
+
+ newmask = kzalloc(sizeof(*newmask), GFP_KERNEL);
+ if (!newmask)
+ return ERR_PTR(-ENOMEM);
+
+ fl_mask_copy(newmask, mask);
+
+ err = fl_init_mask_hashtable(newmask);
+ if (err)
+ goto errout_free;
+
+ fl_init_dissector(newmask);
+
+ INIT_LIST_HEAD_RCU(&newmask->filters);
+
+ err = rhashtable_insert_fast(&head->ht, &newmask->ht_node,
+ mask_ht_params);
+ if (err)
+ goto errout_destroy;
+
+ list_add_tail_rcu(&newmask->list, &head->masks);
+
+ return newmask;
+
+errout_destroy:
+ rhashtable_destroy(&newmask->ht);
+errout_free:
+ kfree(newmask);
+
+ return ERR_PTR(err);
}
static int fl_check_assign_mask(struct cls_fl_head *head,
+ struct cls_fl_filter *fnew,
+ struct cls_fl_filter *fold,
struct fl_flow_mask *mask)
{
- int err;
+ struct fl_flow_mask *newmask;
- if (head->mask_assigned) {
- if (!fl_mask_eq(&head->mask, mask))
+ fnew->mask = rhashtable_lookup_fast(&head->ht, mask, mask_ht_params);
+ if (!fnew->mask) {
+ if (fold)
return -EINVAL;
- else
- return 0;
- }
- /* Mask is not assigned yet. So assign it and init hashtable
- * according to that.
- */
- err = fl_init_hashtable(head, mask);
- if (err)
- return err;
- memcpy(&head->mask, mask, sizeof(head->mask));
- head->mask_assigned = true;
+ newmask = fl_create_new_mask(head, mask);
+ if (IS_ERR(newmask))
+ return PTR_ERR(newmask);
- fl_init_dissector(head, mask);
+ fnew->mask = newmask;
+ } else if (fold && fold->mask == fnew->mask) {
+ return -EINVAL;
+ }
return 0;
}
@@ -924,30 +1001,26 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
if (err)
goto errout_idr;
- err = fl_check_assign_mask(head, &mask);
+ err = fl_check_assign_mask(head, fnew, fold, &mask);
if (err)
goto errout_idr;
if (!tc_skip_sw(fnew->flags)) {
- if (!fold && fl_lookup(head, &fnew->mkey)) {
+ if (!fold && fl_lookup(fnew->mask, &fnew->mkey)) {
err = -EEXIST;
- goto errout_idr;
+ goto errout_mask;
}
- err = rhashtable_insert_fast(&head->ht, &fnew->ht_node,
- head->ht_params);
+ err = rhashtable_insert_fast(&fnew->mask->ht, &fnew->ht_node,
+ fnew->mask->filter_ht_params);
if (err)
- goto errout_idr;
+ goto errout_mask;
}
if (!tc_skip_hw(fnew->flags)) {
- err = fl_hw_replace_filter(tp,
- &head->dissector,
- &mask.key,
- fnew,
- extack);
+ err = fl_hw_replace_filter(tp, fnew, extack);
if (err)
- goto errout_idr;
+ goto errout_mask;
}
if (!tc_in_hw(fnew->flags))
@@ -955,8 +1028,9 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
if (fold) {
if (!tc_skip_sw(fold->flags))
- rhashtable_remove_fast(&head->ht, &fold->ht_node,
- head->ht_params);
+ rhashtable_remove_fast(&fold->mask->ht,
+ &fold->ht_node,
+ fold->mask->filter_ht_params);
if (!tc_skip_hw(fold->flags))
fl_hw_destroy_filter(tp, fold, NULL);
}
@@ -970,12 +1044,15 @@ static int fl_change(struct net *net, struct sk_buff *in_skb,
tcf_exts_get_net(&fold->exts);
call_rcu(&fold->rcu, fl_destroy_filter);
} else {
- list_add_tail_rcu(&fnew->list, &head->filters);
+ list_add_tail_rcu(&fnew->list, &fnew->mask->filters);
}
kfree(tb);
return 0;
+errout_mask:
+ fl_mask_put(head, fnew->mask, false);
+
errout_idr:
if (fnew->handle)
idr_remove(&head->handle_idr, fnew->handle);
@@ -994,10 +1071,10 @@ static int fl_delete(struct tcf_proto *tp, void *arg, bool *last,
struct cls_fl_filter *f = arg;
if (!tc_skip_sw(f->flags))
- rhashtable_remove_fast(&head->ht, &f->ht_node,
- head->ht_params);
+ rhashtable_remove_fast(&f->mask->ht, &f->ht_node,
+ f->mask->filter_ht_params);
__fl_delete(tp, f, extack);
- *last = list_empty(&head->filters);
+ *last = list_empty(&head->masks);
return 0;
}
@@ -1005,16 +1082,19 @@ static void fl_walk(struct tcf_proto *tp, struct tcf_walker *arg)
{
struct cls_fl_head *head = rtnl_dereference(tp->root);
struct cls_fl_filter *f;
-
- list_for_each_entry_rcu(f, &head->filters, list) {
- if (arg->count < arg->skip)
- goto skip;
- if (arg->fn(tp, f, arg) < 0) {
- arg->stop = 1;
- break;
- }
+ struct fl_flow_mask *mask;
+
+ list_for_each_entry_rcu(mask, &head->masks, list) {
+ list_for_each_entry_rcu(f, &mask->filters, list) {
+ if (arg->count < arg->skip)
+ goto skip;
+ if (arg->fn(tp, f, arg) < 0) {
+ arg->stop = 1;
+ break;
+ }
skip:
- arg->count++;
+ arg->count++;
+ }
}
}
@@ -1150,7 +1230,6 @@ static int fl_dump_key_flags(struct sk_buff *skb, u32 flags_key, u32 flags_mask)
static int fl_dump(struct net *net, struct tcf_proto *tp, void *fh,
struct sk_buff *skb, struct tcmsg *t)
{
- struct cls_fl_head *head = rtnl_dereference(tp->root);
struct cls_fl_filter *f = fh;
struct nlattr *nest;
struct fl_flow_key *key, *mask;
@@ -1169,7 +1248,7 @@ static int fl_dump(struct net *net, struct tcf_proto *tp, void *fh,
goto nla_put_failure;
key = &f->key;
- mask = &head->mask.key;
+ mask = &f->mask->key;
if (mask->indev_ifindex) {
struct net_device *dev;
--
2.7.4
^ permalink raw reply related
* Re: [PATCH bpf-next] bpf: remove tracepoints from bpf core
From: Daniel Borkmann @ 2018-04-30 8:20 UTC (permalink / raw)
To: David Miller, ast; +Cc: ebiggers3, netdev, kernel-team
In-Reply-To: <20180429.003739.873200779196020197.davem@davemloft.net>
On 04/29/2018 06:37 AM, David Miller wrote:
> From: Alexei Starovoitov <ast@kernel.org>
> Date: Sat, 28 Apr 2018 19:56:37 -0700
>
>> tracepoints to bpf core were added as a way to provide introspection
>> to bpf programs and maps, but after some time it became clear that
>> this approach is inadequate, so prog_id, map_id and corresponding
>> get_next_id, get_fd_by_id, get_info_by_fd, prog_query APIs were
>> introduced and fully adopted by bpftool and other applications.
>> The tracepoints in bpf core started to rot and causing syzbot warnings:
>> WARNING: CPU: 0 PID: 3008 at kernel/trace/trace_event_perf.c:274
>> Kernel panic - not syncing: panic_on_warn set ...
>> perf_trace_bpf_map_keyval+0x260/0xbd0 include/trace/events/bpf.h:228
>> trace_bpf_map_update_elem include/trace/events/bpf.h:274 [inline]
>> map_update_elem kernel/bpf/syscall.c:597 [inline]
>> SYSC_bpf kernel/bpf/syscall.c:1478 [inline]
>> Hence this patch deletes tracepoints in bpf core.
>>
>> Reported-by: Eric Biggers <ebiggers3@gmail.com>
>> Reported-by: syzbot <bot+a9dbb3c3e64b62536a4bc5ee7bbd4ca627566188@syzkaller.appspotmail.com>
>> Signed-off-by: Alexei Starovoitov <ast@kernel.org>
>
> Acked-by: David S. Miller <davem@davemloft.net>
Fully agree, applied to bpf-next, thanks!
^ permalink raw reply
* Re: [PATCH] DT: net: can: rcar_canfd: document R8A77980 bindings
From: Marc Kleine-Budde @ 2018-04-30 8:18 UTC (permalink / raw)
To: Sergei Shtylyov, Rob Herring, linux-can, netdev, devicetree
Cc: Wolfgang Grandegger, Mark Rutland, linux-renesas-soc
In-Reply-To: <556f1f23-19bc-e7bd-47b1-f22e086e2b69@cogentembedded.com>
[-- Attachment #1.1: Type: text/plain, Size: 518 bytes --]
On 04/27/2018 08:53 PM, Sergei Shtylyov wrote:
> Document the R-Car V3H (R8A77980) SoC support in the R-Car CAN-FD bindings.
>
> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Applied to linux-can.
tnx,
Marc
--
Pengutronix e.K. | Marc Kleine-Budde |
Industrial Linux Solutions | Phone: +49-231-2826-924 |
Vertretung West/Dortmund | Fax: +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686 | http://www.pengutronix.de |
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply
* Re: [PATCH] DT: net: can: rcar_canfd: document R8A77970 bindings
From: Marc Kleine-Budde @ 2018-04-30 8:16 UTC (permalink / raw)
To: Ramesh Shanmugasundaram, Sergei Shtylyov, Rob Herring,
linux-can@vger.kernel.org, netdev@vger.kernel.org,
devicetree@vger.kernel.org
Cc: Wolfgang Grandegger, Mark Rutland,
linux-renesas-soc@vger.kernel.org
In-Reply-To: <TY2PR01MB1962E2336D20499CF27134EBC38D0@TY2PR01MB1962.jpnprd01.prod.outlook.com>
[-- Attachment #1.1: Type: text/plain, Size: 746 bytes --]
On 04/27/2018 01:33 PM, Ramesh Shanmugasundaram wrote:
> Hello Sergei,
>
> Thanks for your patch.
>
>> Subject: [PATCH] DT: net: can: rcar_canfd: document R8A77970 bindings
>>
>> Document the R-Car V3M (R8A77970) SoC support in the R-Car CAN-FD
>> bindings.
>>
>> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>
> Reviewed-by: Ramesh Shanmugasundaram <ramesh.shanmugasundaram@bp.renesas.com>
Applied to linux-can.
tnx,
Marc
--
Pengutronix e.K. | Marc Kleine-Budde |
Industrial Linux Solutions | Phone: +49-231-2826-924 |
Vertretung West/Dortmund | Fax: +49-5121-206917-5555 |
Amtsgericht Hildesheim, HRA 2686 | http://www.pengutronix.de |
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
^ permalink raw reply
* Re: [PATCH] DT: net: can: rcar_canfd: document R8A77980 bindings
From: Simon Horman @ 2018-04-30 7:37 UTC (permalink / raw)
To: Sergei Shtylyov
Cc: Marc Kleine-Budde, Rob Herring, linux-can, netdev, devicetree,
Wolfgang Grandegger, Mark Rutland, linux-renesas-soc
In-Reply-To: <556f1f23-19bc-e7bd-47b1-f22e086e2b69@cogentembedded.com>
On Fri, Apr 27, 2018 at 09:53:33PM +0300, Sergei Shtylyov wrote:
> Document the R-Car V3H (R8A77980) SoC support in the R-Car CAN-FD bindings.
>
> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
>
> ---
> The patch is against the 'linux-can-next.git' repo plus the R8A77970 bindings
> patch posted yesterday. Although I wouldn't object if they're both merged to
> the 'linux-can.git' repo instead. :-)
>
> Documentation/devicetree/bindings/net/can/rcar_canfd.txt | 1 +
> 1 file changed, 1 insertion(+)
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
>
> Index: linux-can-next/Documentation/devicetree/bindings/net/can/rcar_canfd.txt
> ===================================================================
> --- linux-can-next.orig/Documentation/devicetree/bindings/net/can/rcar_canfd.txt
> +++ linux-can-next/Documentation/devicetree/bindings/net/can/rcar_canfd.txt
> @@ -7,6 +7,7 @@ Required properties:
> - "renesas,r8a7795-canfd" for R8A7795 (R-Car H3) compatible controller.
> - "renesas,r8a7796-canfd" for R8A7796 (R-Car M3) compatible controller.
> - "renesas,r8a77970-canfd" for R8A77970 (R-Car V3M) compatible controller.
> + - "renesas,r8a77980-canfd" for R8A77980 (R-Car V3H) compatible controller.
>
> When compatible with the generic version, nodes must list the
> SoC-specific version corresponding to the platform first, followed by the
>
^ permalink raw reply
* Re: [PATCH net-next v9 3/4] virtio_net: Extend virtio to use VF datapath when available
From: Jiri Pirko @ 2018-04-30 7:20 UTC (permalink / raw)
To: Samudrala, Sridhar
Cc: mst, stephen, davem, netdev, virtualization, virtio-dev,
jesse.brandeburg, alexander.h.duyck, kubakici, jasowang,
loseweigh, aaron.f.brown
In-Reply-To: <638e52c7-64db-9d8d-7530-f6b000d3e292@intel.com>
Mon, Apr 30, 2018 at 06:16:58AM CEST, sridhar.samudrala@intel.com wrote:
>On 4/28/2018 2:42 AM, Jiri Pirko wrote:
>> Fri, Apr 27, 2018 at 07:06:59PM CEST,sridhar.samudrala@intel.com wrote:
>> > This patch enables virtio_net to switch over to a VF datapath when a VF
>> > netdev is present with the same MAC address. It allows live migration
>> > of a VM with a direct attached VF without the need to setup a bond/team
>> > between a VF and virtio net device in the guest.
>> >
>> > The hypervisor needs to enable only one datapath at any time so that
>> > packets don't get looped back to the VM over the other datapath. When a VF
>> > is plugged, the virtio datapath link state can be marked as down. The
>> > hypervisor needs to unplug the VF device from the guest on the source host
>> > and reset the MAC filter of the VF to initiate failover of datapath to
>> > virtio before starting the migration. After the migration is completed,
>> > the destination hypervisor sets the MAC filter on the VF and plugs it back
>> > to the guest to switch over to VF datapath.
>> >
>> > It uses the generic failover framework that provides 2 functions to create
>> > and destroy a master failover netdev. When STANDBY feature is enabled, an
>> > additional netdev(failover netdev) is created that acts as a master device
>> > and tracks the state of the 2 lower netdevs. The original virtio_net netdev
>> > is marked as 'standby' netdev and a passthru device with the same MAC is
>> > registered as 'primary' netdev.
>> >
>> > This patch is based on the discussion initiated by Jesse on this thread.
>> > https://marc.info/?l=linux-virtualization&m=151189725224231&w=2
>> >
>> When I enabled the standby feature (hardcoded), I have 2 netdevices now:
>> 4: ens3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UP group default qlen 1000
>> link/ether 52:54:00:b2:a7:f1 brd ff:ff:ff:ff:ff:ff
>> inet6 fe80::5054:ff:feb2:a7f1/64 scope link
>> valid_lft forever preferred_lft forever
>> 5: ens3n_sby: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
>> link/ether 52:54:00:b2:a7:f1 brd ff:ff:ff:ff:ff:ff
>> inet6 fe80::5054:ff:feb2:a7f1/64 scope link
>> valid_lft forever preferred_lft forever
>>
>> However, it seems to confuse my initscripts on Fedora:
>> [root@test1 ~]# ifup ens3
>> ./network-functions: line 78: [: /etc/dhcp/dhclient-ens3: binary operator expected
>> ./network-functions: line 80: [: /etc/dhclient-ens3: binary operator expected
>> ./network-functions: line 69: [: /var/lib/dhclient/dhclient-ens3: binary operator expected
>>
>> Determining IP information for ens3
>> ens3n_sby...Cannot find device "ens3n_sby.pid"
>> Cannot find device "ens3n_sby.lease"
>> failed.
>>
>> I tried to change the standby device mac:
>> ip link set ens3n_sby addr 52:54:00:b2:a7:f2
>> [root@test1 ~]# ifup ens3
>>
>> Determining IP information for ens3... done.
>> [root@test1 ~]#
>>
>> But now the network does not work. I think that the mac change on
>> standby device should be probably refused, no?
>
>Yes. we should block changing standby device mac.
>
>> When I change the mac back, all works fine.
>
>This is strange. So you had to change the standby device mac twice to
>get dhcp working.
Yes. First to some other mac to not to confuse initscripts, second back
to the "failover mac" in order to make frames go through (I'm guessing
that virtio_net also has mac filter for incoming frames).
>
>I do see NetworkManager trying to get dhcp address on standby device, but
Not using NM. Just good old Fedora initscripts.
>i don't see any issue with connectivity. To be totally transparent, we
>need to only expose one netdev.
Yep.
>
>>
>> Now I try to change mac of the failover master:
>> [root@test1 ~]# ip link set ens3 addr 52:54:00:b2:a7:f3
>> RTNETLINK answers: Operation not supported
>>
>> That I did expect to work. I would expect this would change the mac of
>> the master and both standby and primary slaves.
>
>If a VF is untrusted, a VM will not able to change its MAC and moreover
Note that at this point, I have no VF. So I'm not sure why you mention
that.
>in this mode we are assuming that the hypervisor has assigned the MAC and
>guest is not expected to change the MAC.
Wait, for ordinary old-fashioned virtio_net, as a VM user, I can change
mac and all works fine. How is this different? Change mac on "failover
instance" should work and should propagate the mac down to its slaves.
>
>For the initial implementation, i would propose not allowing the guest to
>change the MAC of failover or standby dev.
I see no reason for such restriction.
>
>
>>
>> Now I tried to add a primary pci device. I don't have any fancy VF on my
>> test setup, but I expected the good old 8139cp to work:
>> [root@test1 ~]# ethtool -i ens9
>> driver: 8139cp
>> ....
>> [root@test1 ~]# ip link set ens9 addr 52:54:00:b2:a7:f1
>>
>> I see no message in dmesg, so I guess the failover module did not
>> enslave this netdev. The mac change is not monitored. I would expect
>> that it is and whenever a device changes mac to the failover one, it
>> should be enslaved and whenever it changes mac back to something else,
>> it should be released - the primary one ofcourse.
>
>Sure. that may be the best way to handle the guest changing the primary
>netdev's mac.
Yep.
>
>>
>>
>>
>> [...]
>>
>> > +static int virtnet_get_phys_port_name(struct net_device *dev, char *buf,
>> > + size_t len)
>> > +{
>> > + struct virtnet_info *vi = netdev_priv(dev);
>> > + int ret;
>> > +
>> > + if (!virtio_has_feature(vi->vdev, VIRTIO_NET_F_STANDBY))
>> > + return -EOPNOTSUPP;
>> > +
>> > + ret = snprintf(buf, len, "_sby");
>> please avoid the "_".
>>
>> [...]
>
^ permalink raw reply
* Re: [PATCH] DT: net: can: rcar_canfd: document R8A77970 bindings
From: Simon Horman @ 2018-04-30 7:19 UTC (permalink / raw)
To: Sergei Shtylyov
Cc: Marc Kleine-Budde, Rob Herring, linux-can, netdev, devicetree,
Wolfgang Grandegger, Mark Rutland, linux-renesas-soc
In-Reply-To: <7a3d170f-1d3a-a807-9256-15fe1a78ca4e@cogentembedded.com>
On Thu, Apr 26, 2018 at 10:41:14PM +0300, Sergei Shtylyov wrote:
> Document the R-Car V3M (R8A77970) SoC support in the R-Car CAN-FD bindings.
>
> Signed-off-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Reviewed-by: Simon Horman <horms+renesas@verge.net.au>
>
> ---
> The patch is against the 'linux-can-next.git' repo but I wouldn't object if
> it's merged to the 'linux-can.git' repo instead. :-)
>
> Documentation/devicetree/bindings/net/can/rcar_canfd.txt | 1 +
> 1 file changed, 1 insertion(+)
>
> Index: linux-can-next/Documentation/devicetree/bindings/net/can/rcar_canfd.txt
> ===================================================================
> --- linux-can-next.orig/Documentation/devicetree/bindings/net/can/rcar_canfd.txt
> +++ linux-can-next/Documentation/devicetree/bindings/net/can/rcar_canfd.txt
> @@ -6,6 +6,7 @@ Required properties:
> - "renesas,rcar-gen3-canfd" for R-Car Gen3 compatible controller.
> - "renesas,r8a7795-canfd" for R8A7795 (R-Car H3) compatible controller.
> - "renesas,r8a7796-canfd" for R8A7796 (R-Car M3) compatible controller.
> + - "renesas,r8a77970-canfd" for R8A77970 (R-Car V3M) compatible controller.
>
> When compatible with the generic version, nodes must list the
> SoC-specific version corresponding to the platform first, followed by the
>
^ permalink raw reply
* [PATCH v2 2/2] selftests/bpf: update .gitignore files
From: Sirio Balmelli @ 2018-04-30 7:17 UTC (permalink / raw)
To: daniel; +Cc: netdev
Build products polluting Git cache. Add to respective .gitignore files.
Signed-off-by: Sirio Balmelli <sirio@b-ad.ch>
---
tools/bpf/bpftool/.gitignore | 12 ++++++++++++
tools/testing/selftests/bpf/.gitignore | 1 +
2 files changed, 13 insertions(+)
create mode 100644 tools/bpf/bpftool/.gitignore
diff --git a/tools/bpf/bpftool/.gitignore b/tools/bpf/bpftool/.gitignore
new file mode 100644
index 0000000..427243f
--- /dev/null
+++ b/tools/bpf/bpftool/.gitignore
@@ -0,0 +1,12 @@
+bpftool
+cfg.d
+cgroup.d
+common.d
+disasm.d
+FEATURE-DUMP.bpftool
+jit_disasm.d
+json_writer.d
+main.d
+map.d
+prog.d
+xlated_dumper.d
diff --git a/tools/testing/selftests/bpf/.gitignore b/tools/testing/selftests/bpf/.gitignore
index 3e3b3ce..adc8e54 100644
--- a/tools/testing/selftests/bpf/.gitignore
+++ b/tools/testing/selftests/bpf/.gitignore
@@ -16,3 +16,4 @@ test_sock
test_sock_addr
urandom_read
test_btf
+test_sockmap
--
2.7.4
^ permalink raw reply related
* [PATCH v2 1/2] selftests/bpf: Makefile: add includes to fix broken test build
From: Sirio Balmelli @ 2018-04-30 7:17 UTC (permalink / raw)
To: daniel; +Cc: netdev
Build fails with missing 'asm/bitsperlong.h'.
Include this from tools/arch/[arch]/include/uapi
using an architecture-specific include sourced from 'Module.symvers'
(is this legitimate? another idea was to do a `uname -p` but what if
caller was cross-compiling?)
Once the arch-specific include is fixed, a lack of 'asm/byteorder.h'
now rears its ugly head.
The only sane (ie: will not pull in kernel headers) place I can find it
is in $(ROOT)/usr/include - add this to the include.
Note that $(ROOT)/usr/include is generated on kernel build, hence my
assumption that 'Module.symvers' will also be present.
This highlights a philosophical issue - should these tests build:
1. Against the kernel tree only?
In this case I likely did the right thing here.
2. Against the local machine?
Then the proper handling is to source 'arch' from eg `uname -p`
and include /usr/include/x86_64-linux-gnu/asm/byteorder.h instead
of $(ROOT)/usr/include.
Signed-off-by: Sirio Balmelli <sirio@b-ad.ch>
---
tools/testing/selftests/bpf/Makefile | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
index 9d76218..a19c6af 100644
--- a/tools/testing/selftests/bpf/Makefile
+++ b/tools/testing/selftests/bpf/Makefile
@@ -84,8 +84,14 @@ else
CPU ?= generic
endif
-CLANG_FLAGS = -I. -I./include/uapi -I../../../include/uapi \
- -Wno-compare-distinct-pointer-types
+# Assume caller has compiled the kernel; get compilation architecture from Module.symvers
+TOOLS :=../../..
+ROOT :=../../../..
+ARCH := $(shell sed -rne 's|.*(arch/[^/]+).*|\1|p' $(ROOT)/Module.symvers | head -n 1)
+CLANG_FLAGS = -I. -I./include/uapi \
+ -I$(TOOLS)/include/uapi -I$(TOOLS)/$(ARCH)/include/uapi \
+ -I$(ROOT)/usr/include \
+ -Wno-compare-distinct-pointer-types
$(OUTPUT)/test_l4lb_noinline.o: CLANG_FLAGS += -fno-inline
$(OUTPUT)/test_xdp_noinline.o: CLANG_FLAGS += -fno-inline
--
2.7.4
^ permalink raw reply related
* [PATCH v2 0/2] selftests/bpf
From: Sirio Balmelli @ 2018-04-30 7:17 UTC (permalink / raw)
To: daniel; +Cc: netdev
Thank you for reviewing earlier patch :)
Resubmitting with a saner include chain.
I was unsure whether to reply to your mail directly or submit an entirely new
set of patches marked 'v2'; apologies if this breaks the workflow.
Sirio Balmelli (2):
selftests/bpf: Makefile: add includes to fix broken test build
selftests/bpf: update .gitignore files
tools/bpf/bpftool/.gitignore | 12 ++++++++++++
tools/testing/selftests/bpf/.gitignore | 1 +
tools/testing/selftests/bpf/Makefile | 10 ++++++++--
3 files changed, 21 insertions(+), 2 deletions(-)
create mode 100644 tools/bpf/bpftool/.gitignore
--
2.7.4
^ permalink raw reply
* [PATCH V10 net-next 07/14] net/tls: Add generic NIC offload infrastructure
From: Boris Pismenny @ 2018-04-30 7:16 UTC (permalink / raw)
To: davem
Cc: netdev, saeedm, borisp, davejwatson, ktkhai, sergei.shtylyov,
Ilya Lesokhin, Aviad Yehezkel
In-Reply-To: <1525072583-138506-1-git-send-email-borisp@mellanox.com>
From: Ilya Lesokhin <ilyal@mellanox.com>
This patch adds a generic infrastructure to offload TLS crypto to a
network device. It enables the kernel TLS socket to skip encryption
and authentication operations on the transmit side of the data path.
Leaving those computationally expensive operations to the NIC.
The NIC offload infrastructure builds TLS records and pushes them to
the TCP layer just like the SW KTLS implementation and using the same
API.
TCP segmentation is mostly unaffected. Currently the only exception is
that we prevent mixed SKBs where only part of the payload requires
offload. In the future we are likely to add a similar restriction
following a change cipher spec record.
The notable differences between SW KTLS and NIC offloaded TLS
implementations are as follows:
1. The offloaded implementation builds "plaintext TLS record", those
records contain plaintext instead of ciphertext and place holder bytes
instead of authentication tags.
2. The offloaded implementation maintains a mapping from TCP sequence
number to TLS records. Thus given a TCP SKB sent from a NIC offloaded
TLS socket, we can use the tls NIC offload infrastructure to obtain
enough context to encrypt the payload of the SKB.
A TLS record is released when the last byte of the record is ack'ed,
this is done through the new icsk_clean_acked callback.
The infrastructure should be extendable to support various NIC offload
implementations. However it is currently written with the
implementation below in mind:
The NIC assumes that packets from each offloaded stream are sent as
plaintext and in-order. It keeps track of the TLS records in the TCP
stream. When a packet marked for offload is transmitted, the NIC
encrypts the payload in-place and puts authentication tags in the
relevant place holders.
The responsibility for handling out-of-order packets (i.e. TCP
retransmission, qdisc drops) falls on the netdev driver.
The netdev driver keeps track of the expected TCP SN from the NIC's
perspective. If the next packet to transmit matches the expected TCP
SN, the driver advances the expected TCP SN, and transmits the packet
with TLS offload indication.
If the next packet to transmit does not match the expected TCP SN. The
driver calls the TLS layer to obtain the TLS record that includes the
TCP of the packet for transmission. Using this TLS record, the driver
posts a work entry on the transmit queue to reconstruct the NIC TLS
state required for the offload of the out-of-order packet. It updates
the expected TCP SN accordingly and transmits the now in-order packet.
The same queue is used for packet transmission and TLS context
reconstruction to avoid the need for flushing the transmit queue before
issuing the context reconstruction request.
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com>
---
include/net/tls.h | 69 +++-
net/tls/Kconfig | 10 +
net/tls/Makefile | 2 +
net/tls/tls_device.c | 764 ++++++++++++++++++++++++++++++++++++++++++
net/tls/tls_device_fallback.c | 450 +++++++++++++++++++++++++
net/tls/tls_main.c | 42 ++-
6 files changed, 1332 insertions(+), 5 deletions(-)
create mode 100644 net/tls/tls_device.c
create mode 100644 net/tls/tls_device_fallback.c
diff --git a/include/net/tls.h b/include/net/tls.h
index 95a8c60..8c56809 100644
--- a/include/net/tls.h
+++ b/include/net/tls.h
@@ -116,6 +116,37 @@ struct tls_sw_context_rx {
bool decrypted;
};
+struct tls_record_info {
+ struct list_head list;
+ u32 end_seq;
+ int len;
+ int num_frags;
+ skb_frag_t frags[MAX_SKB_FRAGS];
+};
+
+struct tls_offload_context {
+ struct crypto_aead *aead_send;
+ spinlock_t lock; /* protects records list */
+ struct list_head records_list;
+ struct tls_record_info *open_record;
+ struct tls_record_info *retransmit_hint;
+ u64 hint_record_sn;
+ u64 unacked_record_sn;
+
+ struct scatterlist sg_tx_data[MAX_SKB_FRAGS];
+ void (*sk_destruct)(struct sock *sk);
+ u8 driver_state[];
+ /* The TLS layer reserves room for driver specific state
+ * Currently the belief is that there is not enough
+ * driver specific state to justify another layer of indirection
+ */
+#define TLS_DRIVER_STATE_SIZE (max_t(size_t, 8, sizeof(void *)))
+};
+
+#define TLS_OFFLOAD_CONTEXT_SIZE \
+ (ALIGN(sizeof(struct tls_offload_context), sizeof(void *)) + \
+ TLS_DRIVER_STATE_SIZE)
+
enum {
TLS_PENDING_CLOSED_RECORD
};
@@ -195,9 +226,28 @@ ssize_t tls_sw_splice_read(struct socket *sock, loff_t *ppos,
struct pipe_inode_info *pipe,
size_t len, unsigned int flags);
-void tls_sk_destruct(struct sock *sk, struct tls_context *ctx);
-void tls_icsk_clean_acked(struct sock *sk);
+int tls_set_device_offload(struct sock *sk, struct tls_context *ctx);
+int tls_device_sendmsg(struct sock *sk, struct msghdr *msg, size_t size);
+int tls_device_sendpage(struct sock *sk, struct page *page,
+ int offset, size_t size, int flags);
+void tls_device_sk_destruct(struct sock *sk);
+void tls_device_init(void);
+void tls_device_cleanup(void);
+
+struct tls_record_info *tls_get_record(struct tls_offload_context *context,
+ u32 seq, u64 *p_record_sn);
+
+static inline bool tls_record_is_start_marker(struct tls_record_info *rec)
+{
+ return rec->len == 0;
+}
+
+static inline u32 tls_record_start_seq(struct tls_record_info *rec)
+{
+ return rec->end_seq - rec->len;
+}
+void tls_sk_destruct(struct sock *sk, struct tls_context *ctx);
int tls_push_sg(struct sock *sk, struct tls_context *ctx,
struct scatterlist *sg, u16 first_offset,
int flags);
@@ -234,6 +284,13 @@ static inline bool tls_is_pending_open_record(struct tls_context *tls_ctx)
return tls_ctx->pending_open_record_frags;
}
+static inline bool tls_is_sk_tx_device_offloaded(struct sock *sk)
+{
+ return sk_fullsock(sk) &&
+ /* matches smp_store_release in tls_set_device_offload */
+ smp_load_acquire(&sk->sk_destruct) == &tls_device_sk_destruct;
+}
+
static inline void tls_err_abort(struct sock *sk, int err)
{
sk->sk_err = err;
@@ -329,4 +386,12 @@ int tls_proccess_cmsg(struct sock *sk, struct msghdr *msg,
void tls_register_device(struct tls_device *device);
void tls_unregister_device(struct tls_device *device);
+struct sk_buff *tls_validate_xmit_skb(struct sock *sk,
+ struct net_device *dev,
+ struct sk_buff *skb);
+
+int tls_sw_fallback_init(struct sock *sk,
+ struct tls_offload_context *offload_ctx,
+ struct tls_crypto_info *crypto_info);
+
#endif /* _TLS_OFFLOAD_H */
diff --git a/net/tls/Kconfig b/net/tls/Kconfig
index 89b8745a..73f05ec 100644
--- a/net/tls/Kconfig
+++ b/net/tls/Kconfig
@@ -14,3 +14,13 @@ config TLS
encryption handling of the TLS protocol to be done in-kernel.
If unsure, say N.
+
+config TLS_DEVICE
+ bool "Transport Layer Security HW offload"
+ depends on TLS
+ select SOCK_VALIDATE_XMIT
+ default n
+ help
+ Enable kernel support for HW offload of the TLS protocol.
+
+ If unsure, say N.
diff --git a/net/tls/Makefile b/net/tls/Makefile
index a930fd1..4d6b728 100644
--- a/net/tls/Makefile
+++ b/net/tls/Makefile
@@ -5,3 +5,5 @@
obj-$(CONFIG_TLS) += tls.o
tls-y := tls_main.o tls_sw.o
+
+tls-$(CONFIG_TLS_DEVICE) += tls_device.o tls_device_fallback.o
diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c
new file mode 100644
index 0000000..ac45d62
--- /dev/null
+++ b/net/tls/tls_device.c
@@ -0,0 +1,764 @@
+/* Copyright (c) 2018, Mellanox Technologies All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <crypto/aead.h>
+#include <linux/highmem.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <net/dst.h>
+#include <net/inet_connection_sock.h>
+#include <net/tcp.h>
+#include <net/tls.h>
+
+/* device_offload_lock is used to synchronize tls_dev_add
+ * against NETDEV_DOWN notifications.
+ */
+static DECLARE_RWSEM(device_offload_lock);
+
+static void tls_device_gc_task(struct work_struct *work);
+
+static DECLARE_WORK(tls_device_gc_work, tls_device_gc_task);
+static LIST_HEAD(tls_device_gc_list);
+static LIST_HEAD(tls_device_list);
+static DEFINE_SPINLOCK(tls_device_lock);
+
+static void tls_device_free_ctx(struct tls_context *ctx)
+{
+ struct tls_offload_context *offload_ctx = tls_offload_ctx(ctx);
+
+ kfree(offload_ctx);
+ kfree(ctx);
+}
+
+static void tls_device_gc_task(struct work_struct *work)
+{
+ struct tls_context *ctx, *tmp;
+ unsigned long flags;
+ LIST_HEAD(gc_list);
+
+ spin_lock_irqsave(&tls_device_lock, flags);
+ list_splice_init(&tls_device_gc_list, &gc_list);
+ spin_unlock_irqrestore(&tls_device_lock, flags);
+
+ list_for_each_entry_safe(ctx, tmp, &gc_list, list) {
+ struct net_device *netdev = ctx->netdev;
+
+ if (netdev) {
+ netdev->tlsdev_ops->tls_dev_del(netdev, ctx,
+ TLS_OFFLOAD_CTX_DIR_TX);
+ dev_put(netdev);
+ }
+
+ list_del(&ctx->list);
+ tls_device_free_ctx(ctx);
+ }
+}
+
+static void tls_device_queue_ctx_destruction(struct tls_context *ctx)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&tls_device_lock, flags);
+ list_move_tail(&ctx->list, &tls_device_gc_list);
+
+ /* schedule_work inside the spinlock
+ * to make sure tls_device_down waits for that work.
+ */
+ schedule_work(&tls_device_gc_work);
+
+ spin_unlock_irqrestore(&tls_device_lock, flags);
+}
+
+/* We assume that the socket is already connected */
+static struct net_device *get_netdev_for_sock(struct sock *sk)
+{
+ struct dst_entry *dst = sk_dst_get(sk);
+ struct net_device *netdev = NULL;
+
+ if (likely(dst)) {
+ netdev = dst->dev;
+ dev_hold(netdev);
+ }
+
+ dst_release(dst);
+
+ return netdev;
+}
+
+static void destroy_record(struct tls_record_info *record)
+{
+ int nr_frags = record->num_frags;
+ skb_frag_t *frag;
+
+ while (nr_frags-- > 0) {
+ frag = &record->frags[nr_frags];
+ __skb_frag_unref(frag);
+ }
+ kfree(record);
+}
+
+static void delete_all_records(struct tls_offload_context *offload_ctx)
+{
+ struct tls_record_info *info, *temp;
+
+ list_for_each_entry_safe(info, temp, &offload_ctx->records_list, list) {
+ list_del(&info->list);
+ destroy_record(info);
+ }
+
+ offload_ctx->retransmit_hint = NULL;
+}
+
+static void tls_icsk_clean_acked(struct sock *sk, u32 acked_seq)
+{
+ struct tls_context *tls_ctx = tls_get_ctx(sk);
+ struct tls_record_info *info, *temp;
+ struct tls_offload_context *ctx;
+ u64 deleted_records = 0;
+ unsigned long flags;
+
+ if (!tls_ctx)
+ return;
+
+ ctx = tls_offload_ctx(tls_ctx);
+
+ spin_lock_irqsave(&ctx->lock, flags);
+ info = ctx->retransmit_hint;
+ if (info && !before(acked_seq, info->end_seq)) {
+ ctx->retransmit_hint = NULL;
+ list_del(&info->list);
+ destroy_record(info);
+ deleted_records++;
+ }
+
+ list_for_each_entry_safe(info, temp, &ctx->records_list, list) {
+ if (before(acked_seq, info->end_seq))
+ break;
+ list_del(&info->list);
+
+ destroy_record(info);
+ deleted_records++;
+ }
+
+ ctx->unacked_record_sn += deleted_records;
+ spin_unlock_irqrestore(&ctx->lock, flags);
+}
+
+/* At this point, there should be no references on this
+ * socket and no in-flight SKBs associated with this
+ * socket, so it is safe to free all the resources.
+ */
+void tls_device_sk_destruct(struct sock *sk)
+{
+ struct tls_context *tls_ctx = tls_get_ctx(sk);
+ struct tls_offload_context *ctx = tls_offload_ctx(tls_ctx);
+
+ if (ctx->open_record)
+ destroy_record(ctx->open_record);
+
+ delete_all_records(ctx);
+ crypto_free_aead(ctx->aead_send);
+ ctx->sk_destruct(sk);
+ clean_acked_data_disable(inet_csk(sk));
+
+ if (refcount_dec_and_test(&tls_ctx->refcount))
+ tls_device_queue_ctx_destruction(tls_ctx);
+}
+EXPORT_SYMBOL(tls_device_sk_destruct);
+
+static void tls_append_frag(struct tls_record_info *record,
+ struct page_frag *pfrag,
+ int size)
+{
+ skb_frag_t *frag;
+
+ frag = &record->frags[record->num_frags - 1];
+ if (frag->page.p == pfrag->page &&
+ frag->page_offset + frag->size == pfrag->offset) {
+ frag->size += size;
+ } else {
+ ++frag;
+ frag->page.p = pfrag->page;
+ frag->page_offset = pfrag->offset;
+ frag->size = size;
+ ++record->num_frags;
+ get_page(pfrag->page);
+ }
+
+ pfrag->offset += size;
+ record->len += size;
+}
+
+static int tls_push_record(struct sock *sk,
+ struct tls_context *ctx,
+ struct tls_offload_context *offload_ctx,
+ struct tls_record_info *record,
+ struct page_frag *pfrag,
+ int flags,
+ unsigned char record_type)
+{
+ struct tcp_sock *tp = tcp_sk(sk);
+ struct page_frag dummy_tag_frag;
+ skb_frag_t *frag;
+ int i;
+
+ /* fill prepend */
+ frag = &record->frags[0];
+ tls_fill_prepend(ctx,
+ skb_frag_address(frag),
+ record->len - ctx->tx.prepend_size,
+ record_type);
+
+ /* HW doesn't care about the data in the tag, because it fills it. */
+ dummy_tag_frag.page = skb_frag_page(frag);
+ dummy_tag_frag.offset = 0;
+
+ tls_append_frag(record, &dummy_tag_frag, ctx->tx.tag_size);
+ record->end_seq = tp->write_seq + record->len;
+ spin_lock_irq(&offload_ctx->lock);
+ list_add_tail(&record->list, &offload_ctx->records_list);
+ spin_unlock_irq(&offload_ctx->lock);
+ offload_ctx->open_record = NULL;
+ set_bit(TLS_PENDING_CLOSED_RECORD, &ctx->flags);
+ tls_advance_record_sn(sk, &ctx->tx);
+
+ for (i = 0; i < record->num_frags; i++) {
+ frag = &record->frags[i];
+ sg_unmark_end(&offload_ctx->sg_tx_data[i]);
+ sg_set_page(&offload_ctx->sg_tx_data[i], skb_frag_page(frag),
+ frag->size, frag->page_offset);
+ sk_mem_charge(sk, frag->size);
+ get_page(skb_frag_page(frag));
+ }
+ sg_mark_end(&offload_ctx->sg_tx_data[record->num_frags - 1]);
+
+ /* all ready, send */
+ return tls_push_sg(sk, ctx, offload_ctx->sg_tx_data, 0, flags);
+}
+
+static int tls_create_new_record(struct tls_offload_context *offload_ctx,
+ struct page_frag *pfrag,
+ size_t prepend_size)
+{
+ struct tls_record_info *record;
+ skb_frag_t *frag;
+
+ record = kmalloc(sizeof(*record), GFP_KERNEL);
+ if (!record)
+ return -ENOMEM;
+
+ frag = &record->frags[0];
+ __skb_frag_set_page(frag, pfrag->page);
+ frag->page_offset = pfrag->offset;
+ skb_frag_size_set(frag, prepend_size);
+
+ get_page(pfrag->page);
+ pfrag->offset += prepend_size;
+
+ record->num_frags = 1;
+ record->len = prepend_size;
+ offload_ctx->open_record = record;
+ return 0;
+}
+
+static int tls_do_allocation(struct sock *sk,
+ struct tls_offload_context *offload_ctx,
+ struct page_frag *pfrag,
+ size_t prepend_size)
+{
+ int ret;
+
+ if (!offload_ctx->open_record) {
+ if (unlikely(!skb_page_frag_refill(prepend_size, pfrag,
+ sk->sk_allocation))) {
+ sk->sk_prot->enter_memory_pressure(sk);
+ sk_stream_moderate_sndbuf(sk);
+ return -ENOMEM;
+ }
+
+ ret = tls_create_new_record(offload_ctx, pfrag, prepend_size);
+ if (ret)
+ return ret;
+
+ if (pfrag->size > pfrag->offset)
+ return 0;
+ }
+
+ if (!sk_page_frag_refill(sk, pfrag))
+ return -ENOMEM;
+
+ return 0;
+}
+
+static int tls_push_data(struct sock *sk,
+ struct iov_iter *msg_iter,
+ size_t size, int flags,
+ unsigned char record_type)
+{
+ struct tls_context *tls_ctx = tls_get_ctx(sk);
+ struct tls_offload_context *ctx = tls_offload_ctx(tls_ctx);
+ int tls_push_record_flags = flags | MSG_SENDPAGE_NOTLAST;
+ int more = flags & (MSG_SENDPAGE_NOTLAST | MSG_MORE);
+ struct tls_record_info *record = ctx->open_record;
+ struct page_frag *pfrag;
+ size_t orig_size = size;
+ u32 max_open_record_len;
+ int copy, rc = 0;
+ bool done = false;
+ long timeo;
+
+ if (flags &
+ ~(MSG_MORE | MSG_DONTWAIT | MSG_NOSIGNAL | MSG_SENDPAGE_NOTLAST))
+ return -ENOTSUPP;
+
+ if (sk->sk_err)
+ return -sk->sk_err;
+
+ timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
+ rc = tls_complete_pending_work(sk, tls_ctx, flags, &timeo);
+ if (rc < 0)
+ return rc;
+
+ pfrag = sk_page_frag(sk);
+
+ /* TLS_HEADER_SIZE is not counted as part of the TLS record, and
+ * we need to leave room for an authentication tag.
+ */
+ max_open_record_len = TLS_MAX_PAYLOAD_SIZE +
+ tls_ctx->tx.prepend_size;
+ do {
+ rc = tls_do_allocation(sk, ctx, pfrag,
+ tls_ctx->tx.prepend_size);
+ if (rc) {
+ rc = sk_stream_wait_memory(sk, &timeo);
+ if (!rc)
+ continue;
+
+ record = ctx->open_record;
+ if (!record)
+ break;
+handle_error:
+ if (record_type != TLS_RECORD_TYPE_DATA) {
+ /* avoid sending partial
+ * record with type !=
+ * application_data
+ */
+ size = orig_size;
+ destroy_record(record);
+ ctx->open_record = NULL;
+ } else if (record->len > tls_ctx->tx.prepend_size) {
+ goto last_record;
+ }
+
+ break;
+ }
+
+ record = ctx->open_record;
+ copy = min_t(size_t, size, (pfrag->size - pfrag->offset));
+ copy = min_t(size_t, copy, (max_open_record_len - record->len));
+
+ if (copy_from_iter_nocache(page_address(pfrag->page) +
+ pfrag->offset,
+ copy, msg_iter) != copy) {
+ rc = -EFAULT;
+ goto handle_error;
+ }
+ tls_append_frag(record, pfrag, copy);
+
+ size -= copy;
+ if (!size) {
+last_record:
+ tls_push_record_flags = flags;
+ if (more) {
+ tls_ctx->pending_open_record_frags =
+ record->num_frags;
+ break;
+ }
+
+ done = true;
+ }
+
+ if (done || record->len >= max_open_record_len ||
+ (record->num_frags >= MAX_SKB_FRAGS - 1)) {
+ rc = tls_push_record(sk,
+ tls_ctx,
+ ctx,
+ record,
+ pfrag,
+ tls_push_record_flags,
+ record_type);
+ if (rc < 0)
+ break;
+ }
+ } while (!done);
+
+ if (orig_size - size > 0)
+ rc = orig_size - size;
+
+ return rc;
+}
+
+int tls_device_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
+{
+ unsigned char record_type = TLS_RECORD_TYPE_DATA;
+ int rc;
+
+ lock_sock(sk);
+
+ if (unlikely(msg->msg_controllen)) {
+ rc = tls_proccess_cmsg(sk, msg, &record_type);
+ if (rc)
+ goto out;
+ }
+
+ rc = tls_push_data(sk, &msg->msg_iter, size,
+ msg->msg_flags, record_type);
+
+out:
+ release_sock(sk);
+ return rc;
+}
+
+int tls_device_sendpage(struct sock *sk, struct page *page,
+ int offset, size_t size, int flags)
+{
+ struct iov_iter msg_iter;
+ char *kaddr = kmap(page);
+ struct kvec iov;
+ int rc;
+
+ if (flags & MSG_SENDPAGE_NOTLAST)
+ flags |= MSG_MORE;
+
+ lock_sock(sk);
+
+ if (flags & MSG_OOB) {
+ rc = -ENOTSUPP;
+ goto out;
+ }
+
+ iov.iov_base = kaddr + offset;
+ iov.iov_len = size;
+ iov_iter_kvec(&msg_iter, WRITE | ITER_KVEC, &iov, 1, size);
+ rc = tls_push_data(sk, &msg_iter, size,
+ flags, TLS_RECORD_TYPE_DATA);
+ kunmap(page);
+
+out:
+ release_sock(sk);
+ return rc;
+}
+
+struct tls_record_info *tls_get_record(struct tls_offload_context *context,
+ u32 seq, u64 *p_record_sn)
+{
+ u64 record_sn = context->hint_record_sn;
+ struct tls_record_info *info;
+
+ info = context->retransmit_hint;
+ if (!info ||
+ before(seq, info->end_seq - info->len)) {
+ /* if retransmit_hint is irrelevant start
+ * from the beggining of the list
+ */
+ info = list_first_entry(&context->records_list,
+ struct tls_record_info, list);
+ record_sn = context->unacked_record_sn;
+ }
+
+ list_for_each_entry_from(info, &context->records_list, list) {
+ if (before(seq, info->end_seq)) {
+ if (!context->retransmit_hint ||
+ after(info->end_seq,
+ context->retransmit_hint->end_seq)) {
+ context->hint_record_sn = record_sn;
+ context->retransmit_hint = info;
+ }
+ *p_record_sn = record_sn;
+ return info;
+ }
+ record_sn++;
+ }
+
+ return NULL;
+}
+EXPORT_SYMBOL(tls_get_record);
+
+static int tls_device_push_pending_record(struct sock *sk, int flags)
+{
+ struct iov_iter msg_iter;
+
+ iov_iter_kvec(&msg_iter, WRITE | ITER_KVEC, NULL, 0, 0);
+ return tls_push_data(sk, &msg_iter, 0, flags, TLS_RECORD_TYPE_DATA);
+}
+
+int tls_set_device_offload(struct sock *sk, struct tls_context *ctx)
+{
+ u16 nonce_size, tag_size, iv_size, rec_seq_size;
+ struct tls_record_info *start_marker_record;
+ struct tls_offload_context *offload_ctx;
+ struct tls_crypto_info *crypto_info;
+ struct net_device *netdev;
+ char *iv, *rec_seq;
+ struct sk_buff *skb;
+ int rc = -EINVAL;
+ __be64 rcd_sn;
+
+ if (!ctx)
+ goto out;
+
+ if (ctx->priv_ctx_tx) {
+ rc = -EEXIST;
+ goto out;
+ }
+
+ start_marker_record = kmalloc(sizeof(*start_marker_record), GFP_KERNEL);
+ if (!start_marker_record) {
+ rc = -ENOMEM;
+ goto out;
+ }
+
+ offload_ctx = kzalloc(TLS_OFFLOAD_CONTEXT_SIZE, GFP_KERNEL);
+ if (!offload_ctx) {
+ rc = -ENOMEM;
+ goto free_marker_record;
+ }
+
+ crypto_info = &ctx->crypto_send;
+ switch (crypto_info->cipher_type) {
+ case TLS_CIPHER_AES_GCM_128:
+ nonce_size = TLS_CIPHER_AES_GCM_128_IV_SIZE;
+ tag_size = TLS_CIPHER_AES_GCM_128_TAG_SIZE;
+ iv_size = TLS_CIPHER_AES_GCM_128_IV_SIZE;
+ iv = ((struct tls12_crypto_info_aes_gcm_128 *)crypto_info)->iv;
+ rec_seq_size = TLS_CIPHER_AES_GCM_128_REC_SEQ_SIZE;
+ rec_seq =
+ ((struct tls12_crypto_info_aes_gcm_128 *)crypto_info)->rec_seq;
+ break;
+ default:
+ rc = -EINVAL;
+ goto free_offload_ctx;
+ }
+
+ ctx->tx.prepend_size = TLS_HEADER_SIZE + nonce_size;
+ ctx->tx.tag_size = tag_size;
+ ctx->tx.overhead_size = ctx->tx.prepend_size + ctx->tx.tag_size;
+ ctx->tx.iv_size = iv_size;
+ ctx->tx.iv = kmalloc(iv_size + TLS_CIPHER_AES_GCM_128_SALT_SIZE,
+ GFP_KERNEL);
+ if (!ctx->tx.iv) {
+ rc = -ENOMEM;
+ goto free_offload_ctx;
+ }
+
+ memcpy(ctx->tx.iv + TLS_CIPHER_AES_GCM_128_SALT_SIZE, iv, iv_size);
+
+ ctx->tx.rec_seq_size = rec_seq_size;
+ ctx->tx.rec_seq = kmalloc(rec_seq_size, GFP_KERNEL);
+ if (!ctx->tx.rec_seq) {
+ rc = -ENOMEM;
+ goto free_iv;
+ }
+ memcpy(ctx->tx.rec_seq, rec_seq, rec_seq_size);
+
+ rc = tls_sw_fallback_init(sk, offload_ctx, crypto_info);
+ if (rc)
+ goto free_rec_seq;
+
+ /* start at rec_seq - 1 to account for the start marker record */
+ memcpy(&rcd_sn, ctx->tx.rec_seq, sizeof(rcd_sn));
+ offload_ctx->unacked_record_sn = be64_to_cpu(rcd_sn) - 1;
+
+ start_marker_record->end_seq = tcp_sk(sk)->write_seq;
+ start_marker_record->len = 0;
+ start_marker_record->num_frags = 0;
+
+ INIT_LIST_HEAD(&offload_ctx->records_list);
+ list_add_tail(&start_marker_record->list, &offload_ctx->records_list);
+ spin_lock_init(&offload_ctx->lock);
+
+ clean_acked_data_enable(inet_csk(sk), &tls_icsk_clean_acked);
+ ctx->push_pending_record = tls_device_push_pending_record;
+ offload_ctx->sk_destruct = sk->sk_destruct;
+
+ /* TLS offload is greatly simplified if we don't send
+ * SKBs where only part of the payload needs to be encrypted.
+ * So mark the last skb in the write queue as end of record.
+ */
+ skb = tcp_write_queue_tail(sk);
+ if (skb)
+ TCP_SKB_CB(skb)->eor = 1;
+
+ refcount_set(&ctx->refcount, 1);
+
+ /* We support starting offload on multiple sockets
+ * concurrently, so we only need a read lock here.
+ * This lock must precede get_netdev_for_sock to prevent races between
+ * NETDEV_DOWN and setsockopt.
+ */
+ down_read(&device_offload_lock);
+ netdev = get_netdev_for_sock(sk);
+ if (!netdev) {
+ pr_err_ratelimited("%s: netdev not found\n", __func__);
+ rc = -EINVAL;
+ goto release_lock;
+ }
+
+ if (!(netdev->features & NETIF_F_HW_TLS_TX)) {
+ rc = -ENOTSUPP;
+ goto release_netdev;
+ }
+
+ /* Avoid offloading if the device is down
+ * We don't want to offload new flows after
+ * the NETDEV_DOWN event
+ */
+ if (!(netdev->flags & IFF_UP)) {
+ rc = -EINVAL;
+ goto release_netdev;
+ }
+
+ ctx->priv_ctx_tx = offload_ctx;
+ rc = netdev->tlsdev_ops->tls_dev_add(netdev, sk, TLS_OFFLOAD_CTX_DIR_TX,
+ &ctx->crypto_send,
+ tcp_sk(sk)->write_seq);
+ if (rc)
+ goto release_netdev;
+
+ ctx->netdev = netdev;
+
+ spin_lock_irq(&tls_device_lock);
+ list_add_tail(&ctx->list, &tls_device_list);
+ spin_unlock_irq(&tls_device_lock);
+
+ sk->sk_validate_xmit_skb = tls_validate_xmit_skb;
+ /* following this assignment tls_is_sk_tx_device_offloaded
+ * will return true and the context might be accessed
+ * by the netdev's xmit function.
+ */
+ smp_store_release(&sk->sk_destruct,
+ &tls_device_sk_destruct);
+ up_read(&device_offload_lock);
+ goto out;
+
+release_netdev:
+ dev_put(netdev);
+release_lock:
+ up_read(&device_offload_lock);
+ clean_acked_data_disable(inet_csk(sk));
+ crypto_free_aead(offload_ctx->aead_send);
+free_rec_seq:
+ kfree(ctx->tx.rec_seq);
+free_iv:
+ kfree(ctx->tx.iv);
+free_offload_ctx:
+ kfree(offload_ctx);
+ ctx->priv_ctx_tx = NULL;
+free_marker_record:
+ kfree(start_marker_record);
+out:
+ return rc;
+}
+
+static int tls_device_down(struct net_device *netdev)
+{
+ struct tls_context *ctx, *tmp;
+ unsigned long flags;
+ LIST_HEAD(list);
+
+ /* Request a write lock to block new offload attempts */
+ down_write(&device_offload_lock);
+
+ spin_lock_irqsave(&tls_device_lock, flags);
+ list_for_each_entry_safe(ctx, tmp, &tls_device_list, list) {
+ if (ctx->netdev != netdev ||
+ !refcount_inc_not_zero(&ctx->refcount))
+ continue;
+
+ list_move(&ctx->list, &list);
+ }
+ spin_unlock_irqrestore(&tls_device_lock, flags);
+
+ list_for_each_entry_safe(ctx, tmp, &list, list) {
+ netdev->tlsdev_ops->tls_dev_del(netdev, ctx,
+ TLS_OFFLOAD_CTX_DIR_TX);
+ ctx->netdev = NULL;
+ dev_put(netdev);
+ list_del_init(&ctx->list);
+
+ if (refcount_dec_and_test(&ctx->refcount))
+ tls_device_free_ctx(ctx);
+ }
+
+ up_write(&device_offload_lock);
+
+ flush_work(&tls_device_gc_work);
+
+ return NOTIFY_DONE;
+}
+
+static int tls_dev_event(struct notifier_block *this, unsigned long event,
+ void *ptr)
+{
+ struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+
+ if (!(dev->features & NETIF_F_HW_TLS_TX))
+ return NOTIFY_DONE;
+
+ switch (event) {
+ case NETDEV_REGISTER:
+ case NETDEV_FEAT_CHANGE:
+ if (dev->tlsdev_ops &&
+ dev->tlsdev_ops->tls_dev_add &&
+ dev->tlsdev_ops->tls_dev_del)
+ return NOTIFY_DONE;
+ else
+ return NOTIFY_BAD;
+ case NETDEV_DOWN:
+ return tls_device_down(dev);
+ }
+ return NOTIFY_DONE;
+}
+
+static struct notifier_block tls_dev_notifier = {
+ .notifier_call = tls_dev_event,
+};
+
+void __init tls_device_init(void)
+{
+ register_netdevice_notifier(&tls_dev_notifier);
+}
+
+void __exit tls_device_cleanup(void)
+{
+ unregister_netdevice_notifier(&tls_dev_notifier);
+ flush_work(&tls_device_gc_work);
+}
diff --git a/net/tls/tls_device_fallback.c b/net/tls/tls_device_fallback.c
new file mode 100644
index 0000000..748914a
--- /dev/null
+++ b/net/tls/tls_device_fallback.c
@@ -0,0 +1,450 @@
+/* Copyright (c) 2018, Mellanox Technologies All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <net/tls.h>
+#include <crypto/aead.h>
+#include <crypto/scatterwalk.h>
+#include <net/ip6_checksum.h>
+
+static void chain_to_walk(struct scatterlist *sg, struct scatter_walk *walk)
+{
+ struct scatterlist *src = walk->sg;
+ int diff = walk->offset - src->offset;
+
+ sg_set_page(sg, sg_page(src),
+ src->length - diff, walk->offset);
+
+ scatterwalk_crypto_chain(sg, sg_next(src), 0, 2);
+}
+
+static int tls_enc_record(struct aead_request *aead_req,
+ struct crypto_aead *aead, char *aad,
+ char *iv, __be64 rcd_sn,
+ struct scatter_walk *in,
+ struct scatter_walk *out, int *in_len)
+{
+ unsigned char buf[TLS_HEADER_SIZE + TLS_CIPHER_AES_GCM_128_IV_SIZE];
+ struct scatterlist sg_in[3];
+ struct scatterlist sg_out[3];
+ u16 len;
+ int rc;
+
+ len = min_t(int, *in_len, ARRAY_SIZE(buf));
+
+ scatterwalk_copychunks(buf, in, len, 0);
+ scatterwalk_copychunks(buf, out, len, 1);
+
+ *in_len -= len;
+ if (!*in_len)
+ return 0;
+
+ scatterwalk_pagedone(in, 0, 1);
+ scatterwalk_pagedone(out, 1, 1);
+
+ len = buf[4] | (buf[3] << 8);
+ len -= TLS_CIPHER_AES_GCM_128_IV_SIZE;
+
+ tls_make_aad(aad, len - TLS_CIPHER_AES_GCM_128_TAG_SIZE,
+ (char *)&rcd_sn, sizeof(rcd_sn), buf[0]);
+
+ memcpy(iv + TLS_CIPHER_AES_GCM_128_SALT_SIZE, buf + TLS_HEADER_SIZE,
+ TLS_CIPHER_AES_GCM_128_IV_SIZE);
+
+ sg_init_table(sg_in, ARRAY_SIZE(sg_in));
+ sg_init_table(sg_out, ARRAY_SIZE(sg_out));
+ sg_set_buf(sg_in, aad, TLS_AAD_SPACE_SIZE);
+ sg_set_buf(sg_out, aad, TLS_AAD_SPACE_SIZE);
+ chain_to_walk(sg_in + 1, in);
+ chain_to_walk(sg_out + 1, out);
+
+ *in_len -= len;
+ if (*in_len < 0) {
+ *in_len += TLS_CIPHER_AES_GCM_128_TAG_SIZE;
+ /* the input buffer doesn't contain the entire record.
+ * trim len accordingly. The resulting authentication tag
+ * will contain garbage, but we don't care, so we won't
+ * include any of it in the output skb
+ * Note that we assume the output buffer length
+ * is larger then input buffer length + tag size
+ */
+ if (*in_len < 0)
+ len += *in_len;
+
+ *in_len = 0;
+ }
+
+ if (*in_len) {
+ scatterwalk_copychunks(NULL, in, len, 2);
+ scatterwalk_pagedone(in, 0, 1);
+ scatterwalk_copychunks(NULL, out, len, 2);
+ scatterwalk_pagedone(out, 1, 1);
+ }
+
+ len -= TLS_CIPHER_AES_GCM_128_TAG_SIZE;
+ aead_request_set_crypt(aead_req, sg_in, sg_out, len, iv);
+
+ rc = crypto_aead_encrypt(aead_req);
+
+ return rc;
+}
+
+static void tls_init_aead_request(struct aead_request *aead_req,
+ struct crypto_aead *aead)
+{
+ aead_request_set_tfm(aead_req, aead);
+ aead_request_set_ad(aead_req, TLS_AAD_SPACE_SIZE);
+}
+
+static struct aead_request *tls_alloc_aead_request(struct crypto_aead *aead,
+ gfp_t flags)
+{
+ unsigned int req_size = sizeof(struct aead_request) +
+ crypto_aead_reqsize(aead);
+ struct aead_request *aead_req;
+
+ aead_req = kzalloc(req_size, flags);
+ if (aead_req)
+ tls_init_aead_request(aead_req, aead);
+ return aead_req;
+}
+
+static int tls_enc_records(struct aead_request *aead_req,
+ struct crypto_aead *aead, struct scatterlist *sg_in,
+ struct scatterlist *sg_out, char *aad, char *iv,
+ u64 rcd_sn, int len)
+{
+ struct scatter_walk out, in;
+ int rc;
+
+ scatterwalk_start(&in, sg_in);
+ scatterwalk_start(&out, sg_out);
+
+ do {
+ rc = tls_enc_record(aead_req, aead, aad, iv,
+ cpu_to_be64(rcd_sn), &in, &out, &len);
+ rcd_sn++;
+
+ } while (rc == 0 && len);
+
+ scatterwalk_done(&in, 0, 0);
+ scatterwalk_done(&out, 1, 0);
+
+ return rc;
+}
+
+/* Can't use icsk->icsk_af_ops->send_check here because the ip addresses
+ * might have been changed by NAT.
+ */
+static void update_chksum(struct sk_buff *skb, int headln)
+{
+ struct tcphdr *th = tcp_hdr(skb);
+ int datalen = skb->len - headln;
+ const struct ipv6hdr *ipv6h;
+ const struct iphdr *iph;
+
+ /* We only changed the payload so if we are using partial we don't
+ * need to update anything.
+ */
+ if (likely(skb->ip_summed == CHECKSUM_PARTIAL))
+ return;
+
+ skb->ip_summed = CHECKSUM_PARTIAL;
+ skb->csum_start = skb_transport_header(skb) - skb->head;
+ skb->csum_offset = offsetof(struct tcphdr, check);
+
+ if (skb->sk->sk_family == AF_INET6) {
+ ipv6h = ipv6_hdr(skb);
+ th->check = ~csum_ipv6_magic(&ipv6h->saddr, &ipv6h->daddr,
+ datalen, IPPROTO_TCP, 0);
+ } else {
+ iph = ip_hdr(skb);
+ th->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, datalen,
+ IPPROTO_TCP, 0);
+ }
+}
+
+static void complete_skb(struct sk_buff *nskb, struct sk_buff *skb, int headln)
+{
+ skb_copy_header(nskb, skb);
+
+ skb_put(nskb, skb->len);
+ memcpy(nskb->data, skb->data, headln);
+ update_chksum(nskb, headln);
+
+ nskb->destructor = skb->destructor;
+ nskb->sk = skb->sk;
+ skb->destructor = NULL;
+ skb->sk = NULL;
+ refcount_add(nskb->truesize - skb->truesize,
+ &nskb->sk->sk_wmem_alloc);
+}
+
+/* This function may be called after the user socket is already
+ * closed so make sure we don't use anything freed during
+ * tls_sk_proto_close here
+ */
+
+static int fill_sg_in(struct scatterlist *sg_in,
+ struct sk_buff *skb,
+ struct tls_offload_context *ctx,
+ u64 *rcd_sn,
+ s32 *sync_size,
+ int *resync_sgs)
+{
+ int tcp_payload_offset = skb_transport_offset(skb) + tcp_hdrlen(skb);
+ int payload_len = skb->len - tcp_payload_offset;
+ u32 tcp_seq = ntohl(tcp_hdr(skb)->seq);
+ struct tls_record_info *record;
+ unsigned long flags;
+ int remaining;
+ int i;
+
+ spin_lock_irqsave(&ctx->lock, flags);
+ record = tls_get_record(ctx, tcp_seq, rcd_sn);
+ if (!record) {
+ spin_unlock_irqrestore(&ctx->lock, flags);
+ WARN(1, "Record not found for seq %u\n", tcp_seq);
+ return -EINVAL;
+ }
+
+ *sync_size = tcp_seq - tls_record_start_seq(record);
+ if (*sync_size < 0) {
+ int is_start_marker = tls_record_is_start_marker(record);
+
+ spin_unlock_irqrestore(&ctx->lock, flags);
+ /* This should only occur if the relevant record was
+ * already acked. In that case it should be ok
+ * to drop the packet and avoid retransmission.
+ *
+ * There is a corner case where the packet contains
+ * both an acked and a non-acked record.
+ * We currently don't handle that case and rely
+ * on TCP to retranmit a packet that doesn't contain
+ * already acked payload.
+ */
+ if (!is_start_marker)
+ *sync_size = 0;
+ return -EINVAL;
+ }
+
+ remaining = *sync_size;
+ for (i = 0; remaining > 0; i++) {
+ skb_frag_t *frag = &record->frags[i];
+
+ __skb_frag_ref(frag);
+ sg_set_page(sg_in + i, skb_frag_page(frag),
+ skb_frag_size(frag), frag->page_offset);
+
+ remaining -= skb_frag_size(frag);
+
+ if (remaining < 0)
+ sg_in[i].length += remaining;
+ }
+ *resync_sgs = i;
+
+ spin_unlock_irqrestore(&ctx->lock, flags);
+ if (skb_to_sgvec(skb, &sg_in[i], tcp_payload_offset, payload_len) < 0)
+ return -EINVAL;
+
+ return 0;
+}
+
+static void fill_sg_out(struct scatterlist sg_out[3], void *buf,
+ struct tls_context *tls_ctx,
+ struct sk_buff *nskb,
+ int tcp_payload_offset,
+ int payload_len,
+ int sync_size,
+ void *dummy_buf)
+{
+ sg_set_buf(&sg_out[0], dummy_buf, sync_size);
+ sg_set_buf(&sg_out[1], nskb->data + tcp_payload_offset, payload_len);
+ /* Add room for authentication tag produced by crypto */
+ dummy_buf += sync_size;
+ sg_set_buf(&sg_out[2], dummy_buf, TLS_CIPHER_AES_GCM_128_TAG_SIZE);
+}
+
+static struct sk_buff *tls_enc_skb(struct tls_context *tls_ctx,
+ struct scatterlist sg_out[3],
+ struct scatterlist *sg_in,
+ struct sk_buff *skb,
+ s32 sync_size, u64 rcd_sn)
+{
+ int tcp_payload_offset = skb_transport_offset(skb) + tcp_hdrlen(skb);
+ struct tls_offload_context *ctx = tls_offload_ctx(tls_ctx);
+ int payload_len = skb->len - tcp_payload_offset;
+ void *buf, *iv, *aad, *dummy_buf;
+ struct aead_request *aead_req;
+ struct sk_buff *nskb = NULL;
+ int buf_len;
+
+ aead_req = tls_alloc_aead_request(ctx->aead_send, GFP_ATOMIC);
+ if (!aead_req)
+ return NULL;
+
+ buf_len = TLS_CIPHER_AES_GCM_128_SALT_SIZE +
+ TLS_CIPHER_AES_GCM_128_IV_SIZE +
+ TLS_AAD_SPACE_SIZE +
+ sync_size +
+ TLS_CIPHER_AES_GCM_128_TAG_SIZE;
+ buf = kmalloc(buf_len, GFP_ATOMIC);
+ if (!buf)
+ goto free_req;
+
+ iv = buf;
+ memcpy(iv, tls_ctx->crypto_send_aes_gcm_128.salt,
+ TLS_CIPHER_AES_GCM_128_SALT_SIZE);
+ aad = buf + TLS_CIPHER_AES_GCM_128_SALT_SIZE +
+ TLS_CIPHER_AES_GCM_128_IV_SIZE;
+ dummy_buf = aad + TLS_AAD_SPACE_SIZE;
+
+ nskb = alloc_skb(skb_headroom(skb) + skb->len, GFP_ATOMIC);
+ if (!nskb)
+ goto free_buf;
+
+ skb_reserve(nskb, skb_headroom(skb));
+
+ fill_sg_out(sg_out, buf, tls_ctx, nskb, tcp_payload_offset,
+ payload_len, sync_size, dummy_buf);
+
+ if (tls_enc_records(aead_req, ctx->aead_send, sg_in, sg_out, aad, iv,
+ rcd_sn, sync_size + payload_len) < 0)
+ goto free_nskb;
+
+ complete_skb(nskb, skb, tcp_payload_offset);
+
+ /* validate_xmit_skb_list assumes that if the skb wasn't segmented
+ * nskb->prev will point to the skb itself
+ */
+ nskb->prev = nskb;
+
+free_buf:
+ kfree(buf);
+free_req:
+ kfree(aead_req);
+ return nskb;
+free_nskb:
+ kfree_skb(nskb);
+ nskb = NULL;
+ goto free_buf;
+}
+
+static struct sk_buff *tls_sw_fallback(struct sock *sk, struct sk_buff *skb)
+{
+ int tcp_payload_offset = skb_transport_offset(skb) + tcp_hdrlen(skb);
+ struct tls_context *tls_ctx = tls_get_ctx(sk);
+ struct tls_offload_context *ctx = tls_offload_ctx(tls_ctx);
+ int payload_len = skb->len - tcp_payload_offset;
+ struct scatterlist *sg_in, sg_out[3];
+ struct sk_buff *nskb = NULL;
+ int sg_in_max_elements;
+ int resync_sgs = 0;
+ s32 sync_size = 0;
+ u64 rcd_sn;
+
+ /* worst case is:
+ * MAX_SKB_FRAGS in tls_record_info
+ * MAX_SKB_FRAGS + 1 in SKB head and frags.
+ */
+ sg_in_max_elements = 2 * MAX_SKB_FRAGS + 1;
+
+ if (!payload_len)
+ return skb;
+
+ sg_in = kmalloc_array(sg_in_max_elements, sizeof(*sg_in), GFP_ATOMIC);
+ if (!sg_in)
+ goto free_orig;
+
+ sg_init_table(sg_in, sg_in_max_elements);
+ sg_init_table(sg_out, ARRAY_SIZE(sg_out));
+
+ if (fill_sg_in(sg_in, skb, ctx, &rcd_sn, &sync_size, &resync_sgs)) {
+ /* bypass packets before kernel TLS socket option was set */
+ if (sync_size < 0 && payload_len <= -sync_size)
+ nskb = skb_get(skb);
+ goto put_sg;
+ }
+
+ nskb = tls_enc_skb(tls_ctx, sg_out, sg_in, skb, sync_size, rcd_sn);
+
+put_sg:
+ while (resync_sgs)
+ put_page(sg_page(&sg_in[--resync_sgs]));
+ kfree(sg_in);
+free_orig:
+ kfree_skb(skb);
+ return nskb;
+}
+
+struct sk_buff *tls_validate_xmit_skb(struct sock *sk,
+ struct net_device *dev,
+ struct sk_buff *skb)
+{
+ if (dev == tls_get_ctx(sk)->netdev)
+ return skb;
+
+ return tls_sw_fallback(sk, skb);
+}
+
+int tls_sw_fallback_init(struct sock *sk,
+ struct tls_offload_context *offload_ctx,
+ struct tls_crypto_info *crypto_info)
+{
+ const u8 *key;
+ int rc;
+
+ offload_ctx->aead_send =
+ crypto_alloc_aead("gcm(aes)", 0, CRYPTO_ALG_ASYNC);
+ if (IS_ERR(offload_ctx->aead_send)) {
+ rc = PTR_ERR(offload_ctx->aead_send);
+ pr_err_ratelimited("crypto_alloc_aead failed rc=%d\n", rc);
+ offload_ctx->aead_send = NULL;
+ goto err_out;
+ }
+
+ key = ((struct tls12_crypto_info_aes_gcm_128 *)crypto_info)->key;
+
+ rc = crypto_aead_setkey(offload_ctx->aead_send, key,
+ TLS_CIPHER_AES_GCM_128_KEY_SIZE);
+ if (rc)
+ goto free_aead;
+
+ rc = crypto_aead_setauthsize(offload_ctx->aead_send,
+ TLS_CIPHER_AES_GCM_128_TAG_SIZE);
+ if (rc)
+ goto free_aead;
+
+ return 0;
+free_aead:
+ crypto_free_aead(offload_ctx->aead_send);
+err_out:
+ return rc;
+}
diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c
index 545bf34..3aafb87 100644
--- a/net/tls/tls_main.c
+++ b/net/tls/tls_main.c
@@ -54,6 +54,9 @@ enum {
enum {
TLS_BASE,
TLS_SW,
+#ifdef CONFIG_TLS_DEVICE
+ TLS_HW,
+#endif
TLS_HW_RECORD,
TLS_NUM_CONFIG,
};
@@ -280,6 +283,15 @@ static void tls_sk_proto_close(struct sock *sk, long timeout)
tls_sw_free_resources_rx(sk);
}
+#ifdef CONFIG_TLS_DEVICE
+ if (ctx->tx_conf != TLS_HW) {
+#else
+ {
+#endif
+ kfree(ctx);
+ ctx = NULL;
+ }
+
skip_tx_cleanup:
release_sock(sk);
sk_proto_close(sk, timeout);
@@ -442,8 +454,16 @@ static int do_tls_setsockopt_conf(struct sock *sk, char __user *optval,
}
if (tx) {
- rc = tls_set_sw_offload(sk, ctx, 1);
- conf = TLS_SW;
+#ifdef CONFIG_TLS_DEVICE
+ rc = tls_set_device_offload(sk, ctx);
+ conf = TLS_HW;
+ if (rc) {
+#else
+ {
+#endif
+ rc = tls_set_sw_offload(sk, ctx, 1);
+ conf = TLS_SW;
+ }
} else {
rc = tls_set_sw_offload(sk, ctx, 0);
conf = TLS_SW;
@@ -596,6 +616,16 @@ static void build_protos(struct proto prot[TLS_NUM_CONFIG][TLS_NUM_CONFIG],
prot[TLS_SW][TLS_SW].recvmsg = tls_sw_recvmsg;
prot[TLS_SW][TLS_SW].close = tls_sk_proto_close;
+#ifdef CONFIG_TLS_DEVICE
+ prot[TLS_HW][TLS_BASE] = prot[TLS_BASE][TLS_BASE];
+ prot[TLS_HW][TLS_BASE].sendmsg = tls_device_sendmsg;
+ prot[TLS_HW][TLS_BASE].sendpage = tls_device_sendpage;
+
+ prot[TLS_HW][TLS_SW] = prot[TLS_BASE][TLS_SW];
+ prot[TLS_HW][TLS_SW].sendmsg = tls_device_sendmsg;
+ prot[TLS_HW][TLS_SW].sendpage = tls_device_sendpage;
+#endif
+
prot[TLS_HW_RECORD][TLS_HW_RECORD] = *base;
prot[TLS_HW_RECORD][TLS_HW_RECORD].hash = tls_hw_hash;
prot[TLS_HW_RECORD][TLS_HW_RECORD].unhash = tls_hw_unhash;
@@ -630,7 +660,7 @@ static int tls_init(struct sock *sk)
ctx->getsockopt = sk->sk_prot->getsockopt;
ctx->sk_proto_close = sk->sk_prot->close;
- /* Build IPv6 TLS whenever the address of tcpv6_prot changes */
+ /* Build IPv6 TLS whenever the address of tcpv6 _prot changes */
if (ip_ver == TLSV6 &&
unlikely(sk->sk_prot != smp_load_acquire(&saved_tcpv6_prot))) {
mutex_lock(&tcpv6_prot_mutex);
@@ -680,6 +710,9 @@ static int __init tls_register(void)
tls_sw_proto_ops.poll = tls_sw_poll;
tls_sw_proto_ops.splice_read = tls_sw_splice_read;
+#ifdef CONFIG_TLS_DEVICE
+ tls_device_init();
+#endif
tcp_register_ulp(&tcp_tls_ulp_ops);
return 0;
@@ -688,6 +721,9 @@ static int __init tls_register(void)
static void __exit tls_unregister(void)
{
tcp_unregister_ulp(&tcp_tls_ulp_ops);
+#ifdef CONFIG_TLS_DEVICE
+ tls_device_cleanup();
+#endif
}
module_init(tls_register);
--
1.8.3.1
^ permalink raw reply related
* [PATCH V10 net-next 03/14] net: Add Software fallback infrastructure for socket dependent offloads
From: Boris Pismenny @ 2018-04-30 7:16 UTC (permalink / raw)
To: davem
Cc: netdev, saeedm, borisp, davejwatson, ktkhai, sergei.shtylyov,
Ilya Lesokhin
In-Reply-To: <1525072583-138506-1-git-send-email-borisp@mellanox.com>
From: Ilya Lesokhin <ilyal@mellanox.com>
With socket dependent offloads we rely on the netdev to transform
the transmitted packets before sending them to the wire.
When a packet from an offloaded socket is rerouted to a different
device we need to detect it and do the transformation in software.
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
---
include/net/sock.h | 21 +++++++++++++++++++++
net/Kconfig | 3 +++
net/core/dev.c | 4 ++++
3 files changed, 28 insertions(+)
diff --git a/include/net/sock.h b/include/net/sock.h
index 74d725f..3c568b3 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -481,6 +481,11 @@ struct sock {
void (*sk_error_report)(struct sock *sk);
int (*sk_backlog_rcv)(struct sock *sk,
struct sk_buff *skb);
+#ifdef CONFIG_SOCK_VALIDATE_XMIT
+ struct sk_buff* (*sk_validate_xmit_skb)(struct sock *sk,
+ struct net_device *dev,
+ struct sk_buff *skb);
+#endif
void (*sk_destruct)(struct sock *sk);
struct sock_reuseport __rcu *sk_reuseport_cb;
struct rcu_head sk_rcu;
@@ -2332,6 +2337,22 @@ static inline bool sk_fullsock(const struct sock *sk)
return (1 << sk->sk_state) & ~(TCPF_TIME_WAIT | TCPF_NEW_SYN_RECV);
}
+/* Checks if this SKB belongs to an HW offloaded socket
+ * and whether any SW fallbacks are required based on dev.
+ */
+static inline struct sk_buff *sk_validate_xmit_skb(struct sk_buff *skb,
+ struct net_device *dev)
+{
+#ifdef CONFIG_SOCK_VALIDATE_XMIT
+ struct sock *sk = skb->sk;
+
+ if (sk && sk_fullsock(sk) && sk->sk_validate_xmit_skb)
+ skb = sk->sk_validate_xmit_skb(sk, dev, skb);
+#endif
+
+ return skb;
+}
+
/* This helper checks if a socket is a LISTEN or NEW_SYN_RECV
* SYNACK messages can be attached to either ones (depending on SYNCOOKIE)
*/
diff --git a/net/Kconfig b/net/Kconfig
index 6fa1a44..b62089f 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -407,6 +407,9 @@ config GRO_CELLS
bool
default n
+config SOCK_VALIDATE_XMIT
+ bool
+
config NET_DEVLINK
tristate "Network physical/parent device Netlink interface"
help
diff --git a/net/core/dev.c b/net/core/dev.c
index e01c21a..8944e1e 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3112,6 +3112,10 @@ static struct sk_buff *validate_xmit_skb(struct sk_buff *skb, struct net_device
if (unlikely(!skb))
goto out_null;
+ skb = sk_validate_xmit_skb(skb, dev);
+ if (unlikely(!skb))
+ goto out_null;
+
if (netif_needs_gso(skb, features)) {
struct sk_buff *segs;
--
1.8.3.1
^ permalink raw reply related
* [PATCH V10 net-next 08/14] net/mlx5e: Move defines out of ipsec code
From: Boris Pismenny @ 2018-04-30 7:16 UTC (permalink / raw)
To: davem
Cc: netdev, saeedm, borisp, davejwatson, ktkhai, sergei.shtylyov,
Ilya Lesokhin
In-Reply-To: <1525072583-138506-1-git-send-email-borisp@mellanox.com>
From: Ilya Lesokhin <ilyal@mellanox.com>
The defines are not IPSEC specific.
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en.h | 3 +++
drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h | 3 ---
drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c | 5 +----
drivers/net/ethernet/mellanox/mlx5/core/fpga/sdk.h | 2 ++
4 files changed, 6 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 56c748c..66ecb1c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -55,6 +55,9 @@
struct page_pool;
+#define MLX5E_METADATA_ETHER_TYPE (0x8CE4)
+#define MLX5E_METADATA_ETHER_LEN 8
+
#define MLX5_SET_CFG(p, f, v) MLX5_SET(create_flow_group_in, p, f, v)
#define MLX5E_ETH_HARD_MTU (ETH_HLEN + VLAN_HLEN + ETH_FCS_LEN)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
index 1198fc1..93bf10e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ipsec.h
@@ -45,9 +45,6 @@
#define MLX5E_IPSEC_SADB_RX_BITS 10
#define MLX5E_IPSEC_ESN_SCOPE_MID 0x80000000L
-#define MLX5E_METADATA_ETHER_TYPE (0x8CE4)
-#define MLX5E_METADATA_ETHER_LEN 8
-
struct mlx5e_priv;
struct mlx5e_ipsec_sw_stats {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
index 0f5da49..3c4f1f3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/ipsec.c
@@ -43,9 +43,6 @@
#include "fpga/sdk.h"
#include "fpga/core.h"
-#define SBU_QP_QUEUE_SIZE 8
-#define MLX5_FPGA_IPSEC_CMD_TIMEOUT_MSEC (60 * 1000)
-
enum mlx5_fpga_ipsec_cmd_status {
MLX5_FPGA_IPSEC_CMD_PENDING,
MLX5_FPGA_IPSEC_CMD_SEND_FAIL,
@@ -258,7 +255,7 @@ static int mlx5_fpga_ipsec_cmd_wait(void *ctx)
{
struct mlx5_fpga_ipsec_cmd_context *context = ctx;
unsigned long timeout =
- msecs_to_jiffies(MLX5_FPGA_IPSEC_CMD_TIMEOUT_MSEC);
+ msecs_to_jiffies(MLX5_FPGA_CMD_TIMEOUT_MSEC);
int res;
res = wait_for_completion_timeout(&context->complete, timeout);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/sdk.h b/drivers/net/ethernet/mellanox/mlx5/core/fpga/sdk.h
index baa537e..a0573cc 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/sdk.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/sdk.h
@@ -41,6 +41,8 @@
* DOC: Innova SDK
* This header defines the in-kernel API for Innova FPGA client drivers.
*/
+#define SBU_QP_QUEUE_SIZE 8
+#define MLX5_FPGA_CMD_TIMEOUT_MSEC (60 * 1000)
enum mlx5_fpga_access_type {
MLX5_FPGA_ACCESS_TYPE_I2C = 0x0,
--
1.8.3.1
^ permalink raw reply related
* [PATCH V10 net-next 04/14] net: Add TLS offload netdev ops
From: Boris Pismenny @ 2018-04-30 7:16 UTC (permalink / raw)
To: davem
Cc: netdev, saeedm, borisp, davejwatson, ktkhai, sergei.shtylyov,
Ilya Lesokhin, Aviad Yehezkel
In-Reply-To: <1525072583-138506-1-git-send-email-borisp@mellanox.com>
From: Ilya Lesokhin <ilyal@mellanox.com>
Add new netdev ops to add and delete tls context
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com>
---
include/linux/netdevice.h | 24 ++++++++++++++++++++++++
1 file changed, 24 insertions(+)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9e09dd8..a569bf5 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -865,6 +865,26 @@ struct xfrmdev_ops {
};
#endif
+#if IS_ENABLED(CONFIG_TLS_DEVICE)
+enum tls_offload_ctx_dir {
+ TLS_OFFLOAD_CTX_DIR_RX,
+ TLS_OFFLOAD_CTX_DIR_TX,
+};
+
+struct tls_crypto_info;
+struct tls_context;
+
+struct tlsdev_ops {
+ int (*tls_dev_add)(struct net_device *netdev, struct sock *sk,
+ enum tls_offload_ctx_dir direction,
+ struct tls_crypto_info *crypto_info,
+ u32 start_offload_tcp_sn);
+ void (*tls_dev_del)(struct net_device *netdev,
+ struct tls_context *ctx,
+ enum tls_offload_ctx_dir direction);
+};
+#endif
+
struct dev_ifalias {
struct rcu_head rcuhead;
char ifalias[];
@@ -1750,6 +1770,10 @@ struct net_device {
const struct xfrmdev_ops *xfrmdev_ops;
#endif
+#if IS_ENABLED(CONFIG_TLS_DEVICE)
+ const struct tlsdev_ops *tlsdev_ops;
+#endif
+
const struct header_ops *header_ops;
unsigned int flags;
--
1.8.3.1
^ permalink raw reply related
* [PATCH V10 net-next 05/14] net: Add TLS TX offload features
From: Boris Pismenny @ 2018-04-30 7:16 UTC (permalink / raw)
To: davem
Cc: netdev, saeedm, borisp, davejwatson, ktkhai, sergei.shtylyov,
Ilya Lesokhin, Aviad Yehezkel
In-Reply-To: <1525072583-138506-1-git-send-email-borisp@mellanox.com>
From: Ilya Lesokhin <ilyal@mellanox.com>
This patch adds a netdev feature to configure TLS TX offloads.
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Signed-off-by: Aviad Yehezkel <aviadye@mellanox.com>
---
include/linux/netdev_features.h | 2 ++
net/core/ethtool.c | 1 +
2 files changed, 3 insertions(+)
diff --git a/include/linux/netdev_features.h b/include/linux/netdev_features.h
index fe2f3b3..c87c3a3 100644
--- a/include/linux/netdev_features.h
+++ b/include/linux/netdev_features.h
@@ -78,6 +78,7 @@ enum {
NETIF_F_HW_ESP_BIT, /* Hardware ESP transformation offload */
NETIF_F_HW_ESP_TX_CSUM_BIT, /* ESP with TX checksum offload */
NETIF_F_RX_UDP_TUNNEL_PORT_BIT, /* Offload of RX port for UDP tunnels */
+ NETIF_F_HW_TLS_TX_BIT, /* Hardware TLS TX offload */
NETIF_F_GRO_HW_BIT, /* Hardware Generic receive offload */
NETIF_F_HW_TLS_RECORD_BIT, /* Offload TLS record */
@@ -149,6 +150,7 @@ enum {
#define NETIF_F_RX_UDP_TUNNEL_PORT __NETIF_F(RX_UDP_TUNNEL_PORT)
#define NETIF_F_HW_TLS_RECORD __NETIF_F(HW_TLS_RECORD)
#define NETIF_F_GSO_UDP_L4 __NETIF_F(GSO_UDP_L4)
+#define NETIF_F_HW_TLS_TX __NETIF_F(HW_TLS_TX)
#define for_each_netdev_feature(mask_addr, bit) \
for_each_set_bit(bit, (unsigned long *)mask_addr, NETDEV_FEATURE_COUNT)
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
index b849fda..3c6ce8a 100644
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -110,6 +110,7 @@ int ethtool_op_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
[NETIF_F_HW_ESP_TX_CSUM_BIT] = "esp-tx-csum-hw-offload",
[NETIF_F_RX_UDP_TUNNEL_PORT_BIT] = "rx-udp_tunnel-port-offload",
[NETIF_F_HW_TLS_RECORD_BIT] = "tls-hw-record",
+ [NETIF_F_HW_TLS_TX_BIT] = "tls-hw-tx-offload",
};
static const char
--
1.8.3.1
^ permalink raw reply related
* [PATCH V10 net-next 11/14] net/mlx5e: TLS, Add Innova TLS TX offload data path
From: Boris Pismenny @ 2018-04-30 7:16 UTC (permalink / raw)
To: davem
Cc: netdev, saeedm, borisp, davejwatson, ktkhai, sergei.shtylyov,
Ilya Lesokhin
In-Reply-To: <1525072583-138506-1-git-send-email-borisp@mellanox.com>
From: Ilya Lesokhin <ilyal@mellanox.com>
Implement the TLS tx offload data path according to the
requirements of the TLS generic NIC offload infrastructure.
Special metadata ethertype is used to pass information to
the hardware.
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/Makefile | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/en.h | 15 ++
.../mellanox/mlx5/core/en_accel/en_accel.h | 72 ++++++
.../net/ethernet/mellanox/mlx5/core/en_accel/tls.c | 2 +
.../mellanox/mlx5/core/en_accel/tls_rxtx.c | 272 +++++++++++++++++++++
.../mellanox/mlx5/core/en_accel/tls_rxtx.h | 50 ++++
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 2 +
drivers/net/ethernet/mellanox/mlx5/core/en_stats.c | 10 +
drivers/net/ethernet/mellanox/mlx5/core/en_stats.h | 9 +
drivers/net/ethernet/mellanox/mlx5/core/en_tx.c | 37 +--
10 files changed, 455 insertions(+), 16 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index 50872ed..ec785f5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -28,6 +28,6 @@ mlx5_core-$(CONFIG_MLX5_CORE_IPOIB) += ipoib/ipoib.o ipoib/ethtool.o ipoib/ipoib
mlx5_core-$(CONFIG_MLX5_EN_IPSEC) += en_accel/ipsec.o en_accel/ipsec_rxtx.o \
en_accel/ipsec_stats.o
-mlx5_core-$(CONFIG_MLX5_EN_TLS) += en_accel/tls.o
+mlx5_core-$(CONFIG_MLX5_EN_TLS) += en_accel/tls.o en_accel/tls_rxtx.o
CFLAGS_tracepoint.o := -I$(src)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 66ecb1c..ca5b852 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -335,6 +335,7 @@ enum {
MLX5E_SQ_STATE_RECOVERING,
MLX5E_SQ_STATE_IPSEC,
MLX5E_SQ_STATE_AM,
+ MLX5E_SQ_STATE_TLS,
};
struct mlx5e_sq_wqe_info {
@@ -830,6 +831,8 @@ struct mlx5e_profile {
u16 mlx5e_select_queue(struct net_device *dev, struct sk_buff *skb,
void *accel_priv, select_queue_fallback_t fallback);
netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev);
+netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb,
+ struct mlx5e_tx_wqe *wqe, u16 pi);
void mlx5e_completion_event(struct mlx5_core_cq *mcq);
void mlx5e_cq_error_event(struct mlx5_core_cq *mcq, enum mlx5_event event);
@@ -945,6 +948,18 @@ static inline bool mlx5e_tunnel_inner_ft_supported(struct mlx5_core_dev *mdev)
MLX5_CAP_FLOWTABLE_NIC_RX(mdev, ft_field_support.inner_ip_version));
}
+static inline void mlx5e_sq_fetch_wqe(struct mlx5e_txqsq *sq,
+ struct mlx5e_tx_wqe **wqe,
+ u16 *pi)
+{
+ struct mlx5_wq_cyc *wq;
+
+ wq = &sq->wq;
+ *pi = sq->pc & wq->sz_m1;
+ *wqe = mlx5_wq_cyc_get_wqe(wq, *pi);
+ memset(*wqe, 0, sizeof(**wqe));
+}
+
static inline
struct mlx5e_tx_wqe *mlx5e_post_nop(struct mlx5_wq_cyc *wq, u32 sqn, u16 *pc)
{
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
new file mode 100644
index 0000000..68fcb40a
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/en_accel.h
@@ -0,0 +1,72 @@
+/*
+ * Copyright (c) 2018 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef __MLX5E_EN_ACCEL_H__
+#define __MLX5E_EN_ACCEL_H__
+
+#ifdef CONFIG_MLX5_ACCEL
+
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include "en_accel/ipsec_rxtx.h"
+#include "en_accel/tls_rxtx.h"
+#include "en.h"
+
+static inline struct sk_buff *mlx5e_accel_handle_tx(struct sk_buff *skb,
+ struct mlx5e_txqsq *sq,
+ struct net_device *dev,
+ struct mlx5e_tx_wqe **wqe,
+ u16 *pi)
+{
+#ifdef CONFIG_MLX5_EN_TLS
+ if (sq->state & BIT(MLX5E_SQ_STATE_TLS)) {
+ skb = mlx5e_tls_handle_tx_skb(dev, sq, skb, wqe, pi);
+ if (unlikely(!skb))
+ return NULL;
+ }
+#endif
+
+#ifdef CONFIG_MLX5_EN_IPSEC
+ if (sq->state & BIT(MLX5E_SQ_STATE_IPSEC)) {
+ skb = mlx5e_ipsec_handle_tx_skb(dev, *wqe, skb);
+ if (unlikely(!skb))
+ return NULL;
+ }
+#endif
+
+ return skb;
+}
+
+#endif /* CONFIG_MLX5_ACCEL */
+
+#endif /* __MLX5E_EN_ACCEL_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.c
index 38d8810..aa6981c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls.c
@@ -169,5 +169,7 @@ void mlx5e_tls_build_netdev(struct mlx5e_priv *priv)
if (!mlx5_accel_is_tls_device(priv->mdev))
return;
+ netdev->features |= NETIF_F_HW_TLS_TX;
+ netdev->hw_features |= NETIF_F_HW_TLS_TX;
netdev->tlsdev_ops = &mlx5e_tls_ops;
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c
new file mode 100644
index 0000000..49e8d45
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.c
@@ -0,0 +1,272 @@
+/*
+ * Copyright (c) 2018 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#include "en_accel/tls.h"
+#include "en_accel/tls_rxtx.h"
+
+#define SYNDROME_OFFLOAD_REQUIRED 32
+#define SYNDROME_SYNC 33
+
+struct sync_info {
+ u64 rcd_sn;
+ s32 sync_len;
+ int nr_frags;
+ skb_frag_t frags[MAX_SKB_FRAGS];
+};
+
+struct mlx5e_tls_metadata {
+ /* One byte of syndrome followed by 3 bytes of swid */
+ __be32 syndrome_swid;
+ __be16 first_seq;
+ /* packet type ID field */
+ __be16 ethertype;
+} __packed;
+
+static int mlx5e_tls_add_metadata(struct sk_buff *skb, __be32 swid)
+{
+ struct mlx5e_tls_metadata *pet;
+ struct ethhdr *eth;
+
+ if (skb_cow_head(skb, sizeof(struct mlx5e_tls_metadata)))
+ return -ENOMEM;
+
+ eth = (struct ethhdr *)skb_push(skb, sizeof(struct mlx5e_tls_metadata));
+ skb->mac_header -= sizeof(struct mlx5e_tls_metadata);
+ pet = (struct mlx5e_tls_metadata *)(eth + 1);
+
+ memmove(skb->data, skb->data + sizeof(struct mlx5e_tls_metadata),
+ 2 * ETH_ALEN);
+
+ eth->h_proto = cpu_to_be16(MLX5E_METADATA_ETHER_TYPE);
+ pet->syndrome_swid = htonl(SYNDROME_OFFLOAD_REQUIRED << 24) | swid;
+
+ return 0;
+}
+
+static int mlx5e_tls_get_sync_data(struct mlx5e_tls_offload_context *context,
+ u32 tcp_seq, struct sync_info *info)
+{
+ int remaining, i = 0, ret = -EINVAL;
+ struct tls_record_info *record;
+ unsigned long flags;
+ s32 sync_size;
+
+ spin_lock_irqsave(&context->base.lock, flags);
+ record = tls_get_record(&context->base, tcp_seq, &info->rcd_sn);
+
+ if (unlikely(!record))
+ goto out;
+
+ sync_size = tcp_seq - tls_record_start_seq(record);
+ info->sync_len = sync_size;
+ if (unlikely(sync_size < 0)) {
+ if (tls_record_is_start_marker(record))
+ goto done;
+
+ goto out;
+ }
+
+ remaining = sync_size;
+ while (remaining > 0) {
+ info->frags[i] = record->frags[i];
+ __skb_frag_ref(&info->frags[i]);
+ remaining -= skb_frag_size(&info->frags[i]);
+
+ if (remaining < 0)
+ skb_frag_size_add(&info->frags[i], remaining);
+
+ i++;
+ }
+ info->nr_frags = i;
+done:
+ ret = 0;
+out:
+ spin_unlock_irqrestore(&context->base.lock, flags);
+ return ret;
+}
+
+static void mlx5e_tls_complete_sync_skb(struct sk_buff *skb,
+ struct sk_buff *nskb, u32 tcp_seq,
+ int headln, __be64 rcd_sn)
+{
+ struct mlx5e_tls_metadata *pet;
+ u8 syndrome = SYNDROME_SYNC;
+ struct iphdr *iph;
+ struct tcphdr *th;
+ int data_len, mss;
+
+ nskb->dev = skb->dev;
+ skb_reset_mac_header(nskb);
+ skb_set_network_header(nskb, skb_network_offset(skb));
+ skb_set_transport_header(nskb, skb_transport_offset(skb));
+ memcpy(nskb->data, skb->data, headln);
+ memcpy(nskb->data + headln, &rcd_sn, sizeof(rcd_sn));
+
+ iph = ip_hdr(nskb);
+ iph->tot_len = htons(nskb->len - skb_network_offset(nskb));
+ th = tcp_hdr(nskb);
+ data_len = nskb->len - headln;
+ tcp_seq -= data_len;
+ th->seq = htonl(tcp_seq);
+
+ mss = nskb->dev->mtu - (headln - skb_network_offset(nskb));
+ skb_shinfo(nskb)->gso_size = 0;
+ if (data_len > mss) {
+ skb_shinfo(nskb)->gso_size = mss;
+ skb_shinfo(nskb)->gso_segs = DIV_ROUND_UP(data_len, mss);
+ }
+ skb_shinfo(nskb)->gso_type = skb_shinfo(skb)->gso_type;
+
+ pet = (struct mlx5e_tls_metadata *)(nskb->data + sizeof(struct ethhdr));
+ memcpy(pet, &syndrome, sizeof(syndrome));
+ pet->first_seq = htons(tcp_seq);
+
+ /* MLX5 devices don't care about the checksum partial start, offset
+ * and pseudo header
+ */
+ nskb->ip_summed = CHECKSUM_PARTIAL;
+
+ nskb->xmit_more = 1;
+ nskb->queue_mapping = skb->queue_mapping;
+}
+
+static struct sk_buff *
+mlx5e_tls_handle_ooo(struct mlx5e_tls_offload_context *context,
+ struct mlx5e_txqsq *sq, struct sk_buff *skb,
+ struct mlx5e_tx_wqe **wqe,
+ u16 *pi)
+{
+ u32 tcp_seq = ntohl(tcp_hdr(skb)->seq);
+ struct sync_info info;
+ struct sk_buff *nskb;
+ int linear_len = 0;
+ int headln;
+ int i;
+
+ sq->stats.tls_ooo++;
+
+ if (mlx5e_tls_get_sync_data(context, tcp_seq, &info))
+ /* We might get here if a retransmission reaches the driver
+ * after the relevant record is acked.
+ * It should be safe to drop the packet in this case
+ */
+ goto err_out;
+
+ if (unlikely(info.sync_len < 0)) {
+ u32 payload;
+
+ headln = skb_transport_offset(skb) + tcp_hdrlen(skb);
+ payload = skb->len - headln;
+ if (likely(payload <= -info.sync_len))
+ /* SKB payload doesn't require offload
+ */
+ return skb;
+
+ netdev_err(skb->dev,
+ "Can't offload from the middle of an SKB [seq: %X, offload_seq: %X, end_seq: %X]\n",
+ tcp_seq, tcp_seq + payload + info.sync_len,
+ tcp_seq + payload);
+ goto err_out;
+ }
+
+ if (unlikely(mlx5e_tls_add_metadata(skb, context->swid)))
+ goto err_out;
+
+ headln = skb_transport_offset(skb) + tcp_hdrlen(skb);
+ linear_len += headln + sizeof(info.rcd_sn);
+ nskb = alloc_skb(linear_len, GFP_ATOMIC);
+ if (unlikely(!nskb))
+ goto err_out;
+
+ context->expected_seq = tcp_seq + skb->len - headln;
+ skb_put(nskb, linear_len);
+ for (i = 0; i < info.nr_frags; i++)
+ skb_shinfo(nskb)->frags[i] = info.frags[i];
+
+ skb_shinfo(nskb)->nr_frags = info.nr_frags;
+ nskb->data_len = info.sync_len;
+ nskb->len += info.sync_len;
+ sq->stats.tls_resync_bytes += nskb->len;
+ mlx5e_tls_complete_sync_skb(skb, nskb, tcp_seq, headln,
+ cpu_to_be64(info.rcd_sn));
+ mlx5e_sq_xmit(sq, nskb, *wqe, *pi);
+ mlx5e_sq_fetch_wqe(sq, wqe, pi);
+ return skb;
+
+err_out:
+ dev_kfree_skb_any(skb);
+ return NULL;
+}
+
+struct sk_buff *mlx5e_tls_handle_tx_skb(struct net_device *netdev,
+ struct mlx5e_txqsq *sq,
+ struct sk_buff *skb,
+ struct mlx5e_tx_wqe **wqe,
+ u16 *pi)
+{
+ struct mlx5e_tls_offload_context *context;
+ struct tls_context *tls_ctx;
+ u32 expected_seq;
+ int datalen;
+ u32 skb_seq;
+
+ if (!skb->sk || !tls_is_sk_tx_device_offloaded(skb->sk))
+ goto out;
+
+ datalen = skb->len - (skb_transport_offset(skb) + tcp_hdrlen(skb));
+ if (!datalen)
+ goto out;
+
+ tls_ctx = tls_get_ctx(skb->sk);
+ if (unlikely(tls_ctx->netdev != netdev))
+ goto out;
+
+ skb_seq = ntohl(tcp_hdr(skb)->seq);
+ context = mlx5e_get_tls_tx_context(tls_ctx);
+ expected_seq = context->expected_seq;
+
+ if (unlikely(expected_seq != skb_seq)) {
+ skb = mlx5e_tls_handle_ooo(context, sq, skb, wqe, pi);
+ goto out;
+ }
+
+ if (unlikely(mlx5e_tls_add_metadata(skb, context->swid))) {
+ dev_kfree_skb_any(skb);
+ skb = NULL;
+ goto out;
+ }
+
+ context->expected_seq = skb_seq + datalen;
+out:
+ return skb;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.h b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.h
new file mode 100644
index 0000000..405dfd3
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/tls_rxtx.h
@@ -0,0 +1,50 @@
+/*
+ * Copyright (c) 2018 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef __MLX5E_TLS_RXTX_H__
+#define __MLX5E_TLS_RXTX_H__
+
+#ifdef CONFIG_MLX5_EN_TLS
+
+#include <linux/skbuff.h>
+#include "en.h"
+
+struct sk_buff *mlx5e_tls_handle_tx_skb(struct net_device *netdev,
+ struct mlx5e_txqsq *sq,
+ struct sk_buff *skb,
+ struct mlx5e_tx_wqe **wqe,
+ u16 *pi);
+
+#endif /* CONFIG_MLX5_EN_TLS */
+
+#endif /* __MLX5E_TLS_RXTX_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 0450c8c..5f5f97e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1016,6 +1016,8 @@ static int mlx5e_alloc_txqsq(struct mlx5e_channel *c,
INIT_WORK(&sq->recover.recover_work, mlx5e_sq_recover);
if (MLX5_IPSEC_DEV(c->priv->mdev))
set_bit(MLX5E_SQ_STATE_IPSEC, &sq->state);
+ if (mlx5_accel_is_tls_device(c->priv->mdev))
+ set_bit(MLX5E_SQ_STATE_TLS, &sq->state);
param->wq.db_numa_node = cpu_to_node(c->cpu);
err = mlx5_wq_cyc_create(mdev, ¶m->wq, sqc_wq, &sq->wq, &sq->wq_ctrl);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
index b08c944..c04cf2b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.c
@@ -43,6 +43,12 @@
{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tso_inner_packets) },
{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tso_inner_bytes) },
{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_added_vlan_packets) },
+
+#ifdef CONFIG_MLX5_EN_TLS
+ { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_ooo) },
+ { MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, tx_tls_resync_bytes) },
+#endif
+
{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_lro_packets) },
{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_lro_bytes) },
{ MLX5E_DECLARE_STAT(struct mlx5e_sw_stats, rx_removed_vlan_packets) },
@@ -161,6 +167,10 @@ static void mlx5e_grp_sw_update_stats(struct mlx5e_priv *priv)
s->tx_csum_partial_inner += sq_stats->csum_partial_inner;
s->tx_csum_none += sq_stats->csum_none;
s->tx_csum_partial += sq_stats->csum_partial;
+#ifdef CONFIG_MLX5_EN_TLS
+ s->tx_tls_ooo += sq_stats->tls_ooo;
+ s->tx_tls_resync_bytes += sq_stats->tls_resync_bytes;
+#endif
}
}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
index 53111a2..a36e6a8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_stats.h
@@ -93,6 +93,11 @@ struct mlx5e_sw_stats {
u64 rx_cache_waive;
u64 ch_eq_rearm;
+#ifdef CONFIG_MLX5_EN_TLS
+ u64 tx_tls_ooo;
+ u64 tx_tls_resync_bytes;
+#endif
+
/* Special handling counters */
u64 link_down_events_phy;
};
@@ -194,6 +199,10 @@ struct mlx5e_sq_stats {
u64 csum_partial_inner;
u64 added_vlan_packets;
u64 nop;
+#ifdef CONFIG_MLX5_EN_TLS
+ u64 tls_ooo;
+ u64 tls_resync_bytes;
+#endif
/* less likely accessed in data path */
u64 csum_none;
u64 stopped;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index 2029710..486b6bf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -35,12 +35,21 @@
#include <net/dsfield.h>
#include "en.h"
#include "ipoib/ipoib.h"
-#include "en_accel/ipsec_rxtx.h"
+#include "en_accel/en_accel.h"
#include "lib/clock.h"
#define MLX5E_SQ_NOPS_ROOM MLX5_SEND_WQE_MAX_WQEBBS
+
+#ifndef CONFIG_MLX5_EN_TLS
#define MLX5E_SQ_STOP_ROOM (MLX5_SEND_WQE_MAX_WQEBBS +\
MLX5E_SQ_NOPS_ROOM)
+#else
+/* TLS offload requires MLX5E_SQ_STOP_ROOM to have
+ * enough room for a resync SKB, a normal SKB and a NOP
+ */
+#define MLX5E_SQ_STOP_ROOM (2 * MLX5_SEND_WQE_MAX_WQEBBS +\
+ MLX5E_SQ_NOPS_ROOM)
+#endif
static inline void mlx5e_tx_dma_unmap(struct device *pdev,
struct mlx5e_sq_dma *dma)
@@ -325,8 +334,8 @@ static inline void mlx5e_insert_vlan(void *start, struct sk_buff *skb, u16 ihs,
}
}
-static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb,
- struct mlx5e_tx_wqe *wqe, u16 pi)
+netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb,
+ struct mlx5e_tx_wqe *wqe, u16 pi)
{
struct mlx5e_tx_wqe_info *wi = &sq->db.wqe_info[pi];
@@ -399,21 +408,19 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_txqsq *sq, struct sk_buff *skb,
netdev_tx_t mlx5e_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct mlx5e_priv *priv = netdev_priv(dev);
- struct mlx5e_txqsq *sq = priv->txq2sq[skb_get_queue_mapping(skb)];
- struct mlx5_wq_cyc *wq = &sq->wq;
- u16 pi = sq->pc & wq->sz_m1;
- struct mlx5e_tx_wqe *wqe = mlx5_wq_cyc_get_wqe(wq, pi);
+ struct mlx5e_tx_wqe *wqe;
+ struct mlx5e_txqsq *sq;
+ u16 pi;
- memset(wqe, 0, sizeof(*wqe));
+ sq = priv->txq2sq[skb_get_queue_mapping(skb)];
+ mlx5e_sq_fetch_wqe(sq, &wqe, &pi);
-#ifdef CONFIG_MLX5_EN_IPSEC
- if (sq->state & BIT(MLX5E_SQ_STATE_IPSEC)) {
- skb = mlx5e_ipsec_handle_tx_skb(dev, wqe, skb);
- if (unlikely(!skb))
- return NETDEV_TX_OK;
- }
+#ifdef CONFIG_MLX5_ACCEL
+ /* might send skbs and update wqe and pi */
+ skb = mlx5e_accel_handle_tx(skb, sq, dev, &wqe, &pi);
+ if (unlikely(!skb))
+ return NETDEV_TX_OK;
#endif
-
return mlx5e_sq_xmit(sq, skb, wqe, pi);
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH V10 net-next 09/14] net/mlx5: Accel, Add TLS tx offload interface
From: Boris Pismenny @ 2018-04-30 7:16 UTC (permalink / raw)
To: davem
Cc: netdev, saeedm, borisp, davejwatson, ktkhai, sergei.shtylyov,
Ilya Lesokhin
In-Reply-To: <1525072583-138506-1-git-send-email-borisp@mellanox.com>
From: Ilya Lesokhin <ilyal@mellanox.com>
Add routines for manipulating TLS TX offload contexts.
In Innova TLS, TLS contexts are added or deleted
via a command message over the SBU connection.
The HW then sends a response message over the same connection.
Add implementation for Innova TLS (FPGA-based) hardware.
These routines will be used by the TLS offload support in a later patch
mlx5/accel is a middle acceleration layer to allow mlx5e and other ULPs
to work directly with mlx5_core rather than Innova FPGA or other mlx5
acceleration providers.
In the future, when IPSec/TLS or any other acceleration gets integrated
into ConnectX chip, mlx5/accel layer will provide the integrated
acceleration, rather than the Innova one.
Signed-off-by: Ilya Lesokhin <ilyal@mellanox.com>
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/Makefile | 4 +-
.../net/ethernet/mellanox/mlx5/core/accel/tls.c | 71 +++
.../net/ethernet/mellanox/mlx5/core/accel/tls.h | 86 ++++
.../net/ethernet/mellanox/mlx5/core/fpga/core.h | 1 +
drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c | 562 +++++++++++++++++++++
drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h | 68 +++
drivers/net/ethernet/mellanox/mlx5/core/main.c | 11 +
include/linux/mlx5/mlx5_ifc.h | 16 -
include/linux/mlx5/mlx5_ifc_fpga.h | 77 +++
9 files changed, 878 insertions(+), 18 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/accel/tls.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/accel/tls.h
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c
create mode 100644 drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
index c805769..9989e52 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/Makefile
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -8,10 +8,10 @@ mlx5_core-y := main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
fs_counters.o rl.o lag.o dev.o wq.o lib/gid.o lib/clock.o \
diag/fs_tracepoint.o
-mlx5_core-$(CONFIG_MLX5_ACCEL) += accel/ipsec.o
+mlx5_core-$(CONFIG_MLX5_ACCEL) += accel/ipsec.o accel/tls.o
mlx5_core-$(CONFIG_MLX5_FPGA) += fpga/cmd.o fpga/core.o fpga/conn.o fpga/sdk.o \
- fpga/ipsec.o
+ fpga/ipsec.o fpga/tls.o
mlx5_core-$(CONFIG_MLX5_CORE_EN) += en_main.o en_common.o en_fs.o en_ethtool.o \
en_tx.o en_rx.o en_dim.o en_txrx.o en_stats.o vxlan.o \
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.c b/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.c
new file mode 100644
index 0000000..77ac19f
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.c
@@ -0,0 +1,71 @@
+/*
+ * Copyright (c) 2018 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#include <linux/mlx5/device.h>
+
+#include "accel/tls.h"
+#include "mlx5_core.h"
+#include "fpga/tls.h"
+
+int mlx5_accel_tls_add_tx_flow(struct mlx5_core_dev *mdev, void *flow,
+ struct tls_crypto_info *crypto_info,
+ u32 start_offload_tcp_sn, u32 *p_swid)
+{
+ return mlx5_fpga_tls_add_tx_flow(mdev, flow, crypto_info,
+ start_offload_tcp_sn, p_swid);
+}
+
+void mlx5_accel_tls_del_tx_flow(struct mlx5_core_dev *mdev, u32 swid)
+{
+ mlx5_fpga_tls_del_tx_flow(mdev, swid, GFP_KERNEL);
+}
+
+bool mlx5_accel_is_tls_device(struct mlx5_core_dev *mdev)
+{
+ return mlx5_fpga_is_tls_device(mdev);
+}
+
+u32 mlx5_accel_tls_device_caps(struct mlx5_core_dev *mdev)
+{
+ return mlx5_fpga_tls_device_caps(mdev);
+}
+
+int mlx5_accel_tls_init(struct mlx5_core_dev *mdev)
+{
+ return mlx5_fpga_tls_init(mdev);
+}
+
+void mlx5_accel_tls_cleanup(struct mlx5_core_dev *mdev)
+{
+ mlx5_fpga_tls_cleanup(mdev);
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.h b/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.h
new file mode 100644
index 0000000..6f9c9f4
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/accel/tls.h
@@ -0,0 +1,86 @@
+/*
+ * Copyright (c) 2018 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef __MLX5_ACCEL_TLS_H__
+#define __MLX5_ACCEL_TLS_H__
+
+#include <linux/mlx5/driver.h>
+#include <linux/tls.h>
+
+#ifdef CONFIG_MLX5_ACCEL
+
+enum {
+ MLX5_ACCEL_TLS_TX = BIT(0),
+ MLX5_ACCEL_TLS_RX = BIT(1),
+ MLX5_ACCEL_TLS_V12 = BIT(2),
+ MLX5_ACCEL_TLS_V13 = BIT(3),
+ MLX5_ACCEL_TLS_LRO = BIT(4),
+ MLX5_ACCEL_TLS_IPV6 = BIT(5),
+ MLX5_ACCEL_TLS_AES_GCM128 = BIT(30),
+ MLX5_ACCEL_TLS_AES_GCM256 = BIT(31),
+};
+
+struct mlx5_ifc_tls_flow_bits {
+ u8 src_port[0x10];
+ u8 dst_port[0x10];
+ union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits src_ipv4_src_ipv6;
+ union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits dst_ipv4_dst_ipv6;
+ u8 ipv6[0x1];
+ u8 direction_sx[0x1];
+ u8 reserved_at_2[0x1e];
+};
+
+int mlx5_accel_tls_add_tx_flow(struct mlx5_core_dev *mdev, void *flow,
+ struct tls_crypto_info *crypto_info,
+ u32 start_offload_tcp_sn, u32 *p_swid);
+void mlx5_accel_tls_del_tx_flow(struct mlx5_core_dev *mdev, u32 swid);
+bool mlx5_accel_is_tls_device(struct mlx5_core_dev *mdev);
+u32 mlx5_accel_tls_device_caps(struct mlx5_core_dev *mdev);
+int mlx5_accel_tls_init(struct mlx5_core_dev *mdev);
+void mlx5_accel_tls_cleanup(struct mlx5_core_dev *mdev);
+
+#else
+
+static inline int
+mlx5_accel_tls_add_tx_flow(struct mlx5_core_dev *mdev, void *flow,
+ struct tls_crypto_info *crypto_info,
+ u32 start_offload_tcp_sn, u32 *p_swid) { return 0; }
+static inline void mlx5_accel_tls_del_tx_flow(struct mlx5_core_dev *mdev, u32 swid) { }
+static inline bool mlx5_accel_is_tls_device(struct mlx5_core_dev *mdev) { return false; }
+static inline u32 mlx5_accel_tls_device_caps(struct mlx5_core_dev *mdev) { return 0; }
+static inline int mlx5_accel_tls_init(struct mlx5_core_dev *mdev) { return 0; }
+static inline void mlx5_accel_tls_cleanup(struct mlx5_core_dev *mdev) { }
+
+#endif
+
+#endif /* __MLX5_ACCEL_TLS_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/core.h b/drivers/net/ethernet/mellanox/mlx5/core/fpga/core.h
index 82405ed..3e2355c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/core.h
@@ -53,6 +53,7 @@ struct mlx5_fpga_device {
} conn_res;
struct mlx5_fpga_ipsec *ipsec;
+ struct mlx5_fpga_tls *tls;
};
#define mlx5_fpga_dbg(__adev, format, ...) \
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c
new file mode 100644
index 0000000..2104801
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c
@@ -0,0 +1,562 @@
+/*
+ * Copyright (c) 2018 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#include <linux/mlx5/device.h>
+#include "fpga/tls.h"
+#include "fpga/cmd.h"
+#include "fpga/sdk.h"
+#include "fpga/core.h"
+#include "accel/tls.h"
+
+struct mlx5_fpga_tls_command_context;
+
+typedef void (*mlx5_fpga_tls_command_complete)
+ (struct mlx5_fpga_conn *conn, struct mlx5_fpga_device *fdev,
+ struct mlx5_fpga_tls_command_context *ctx,
+ struct mlx5_fpga_dma_buf *resp);
+
+struct mlx5_fpga_tls_command_context {
+ struct list_head list;
+ /* There is no guarantee on the order between the TX completion
+ * and the command response.
+ * The TX completion is going to touch cmd->buf even in
+ * the case of successful transmission.
+ * So instead of requiring separate allocations for cmd
+ * and cmd->buf we've decided to use a reference counter
+ */
+ refcount_t ref;
+ struct mlx5_fpga_dma_buf buf;
+ mlx5_fpga_tls_command_complete complete;
+};
+
+static void
+mlx5_fpga_tls_put_command_ctx(struct mlx5_fpga_tls_command_context *ctx)
+{
+ if (refcount_dec_and_test(&ctx->ref))
+ kfree(ctx);
+}
+
+static void mlx5_fpga_tls_cmd_complete(struct mlx5_fpga_device *fdev,
+ struct mlx5_fpga_dma_buf *resp)
+{
+ struct mlx5_fpga_conn *conn = fdev->tls->conn;
+ struct mlx5_fpga_tls_command_context *ctx;
+ struct mlx5_fpga_tls *tls = fdev->tls;
+ unsigned long flags;
+
+ spin_lock_irqsave(&tls->pending_cmds_lock, flags);
+ ctx = list_first_entry(&tls->pending_cmds,
+ struct mlx5_fpga_tls_command_context, list);
+ list_del(&ctx->list);
+ spin_unlock_irqrestore(&tls->pending_cmds_lock, flags);
+ ctx->complete(conn, fdev, ctx, resp);
+}
+
+static void mlx5_fpga_cmd_send_complete(struct mlx5_fpga_conn *conn,
+ struct mlx5_fpga_device *fdev,
+ struct mlx5_fpga_dma_buf *buf,
+ u8 status)
+{
+ struct mlx5_fpga_tls_command_context *ctx =
+ container_of(buf, struct mlx5_fpga_tls_command_context, buf);
+
+ mlx5_fpga_tls_put_command_ctx(ctx);
+
+ if (unlikely(status))
+ mlx5_fpga_tls_cmd_complete(fdev, NULL);
+}
+
+static void mlx5_fpga_tls_cmd_send(struct mlx5_fpga_device *fdev,
+ struct mlx5_fpga_tls_command_context *cmd,
+ mlx5_fpga_tls_command_complete complete)
+{
+ struct mlx5_fpga_tls *tls = fdev->tls;
+ unsigned long flags;
+ int ret;
+
+ refcount_set(&cmd->ref, 2);
+ cmd->complete = complete;
+ cmd->buf.complete = mlx5_fpga_cmd_send_complete;
+
+ spin_lock_irqsave(&tls->pending_cmds_lock, flags);
+ /* mlx5_fpga_sbu_conn_sendmsg is called under pending_cmds_lock
+ * to make sure commands are inserted to the tls->pending_cmds list
+ * and the command QP in the same order.
+ */
+ ret = mlx5_fpga_sbu_conn_sendmsg(tls->conn, &cmd->buf);
+ if (likely(!ret))
+ list_add_tail(&cmd->list, &tls->pending_cmds);
+ else
+ complete(tls->conn, fdev, cmd, NULL);
+ spin_unlock_irqrestore(&tls->pending_cmds_lock, flags);
+}
+
+/* Start of context identifiers range (inclusive) */
+#define SWID_START 0
+/* End of context identifiers range (exclusive) */
+#define SWID_END BIT(24)
+
+static int mlx5_fpga_tls_alloc_swid(struct idr *idr, spinlock_t *idr_spinlock,
+ void *ptr)
+{
+ int ret;
+
+ /* TLS metadata format is 1 byte for syndrome followed
+ * by 3 bytes of swid (software ID)
+ * swid must not exceed 3 bytes.
+ * See tls_rxtx.c:insert_pet() for details
+ */
+ BUILD_BUG_ON((SWID_END - 1) & 0xFF000000);
+
+ idr_preload(GFP_KERNEL);
+ spin_lock_irq(idr_spinlock);
+ ret = idr_alloc(idr, ptr, SWID_START, SWID_END, GFP_ATOMIC);
+ spin_unlock_irq(idr_spinlock);
+ idr_preload_end();
+
+ return ret;
+}
+
+static void mlx5_fpga_tls_release_swid(struct idr *idr,
+ spinlock_t *idr_spinlock, u32 swid)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(idr_spinlock, flags);
+ idr_remove(idr, swid);
+ spin_unlock_irqrestore(idr_spinlock, flags);
+}
+
+struct mlx5_teardown_stream_context {
+ struct mlx5_fpga_tls_command_context cmd;
+ u32 swid;
+};
+
+static void
+mlx5_fpga_tls_teardown_completion(struct mlx5_fpga_conn *conn,
+ struct mlx5_fpga_device *fdev,
+ struct mlx5_fpga_tls_command_context *cmd,
+ struct mlx5_fpga_dma_buf *resp)
+{
+ struct mlx5_teardown_stream_context *ctx =
+ container_of(cmd, struct mlx5_teardown_stream_context, cmd);
+
+ if (resp) {
+ u32 syndrome = MLX5_GET(tls_resp, resp->sg[0].data, syndrome);
+
+ if (syndrome)
+ mlx5_fpga_err(fdev,
+ "Teardown stream failed with syndrome = %d",
+ syndrome);
+ else
+ mlx5_fpga_tls_release_swid(&fdev->tls->tx_idr,
+ &fdev->tls->idr_spinlock,
+ ctx->swid);
+ }
+ mlx5_fpga_tls_put_command_ctx(cmd);
+}
+
+static void mlx5_fpga_tls_flow_to_cmd(void *flow, void *cmd)
+{
+ memcpy(MLX5_ADDR_OF(tls_cmd, cmd, src_port), flow,
+ MLX5_BYTE_OFF(tls_flow, ipv6));
+
+ MLX5_SET(tls_cmd, cmd, ipv6, MLX5_GET(tls_flow, flow, ipv6));
+ MLX5_SET(tls_cmd, cmd, direction_sx,
+ MLX5_GET(tls_flow, flow, direction_sx));
+}
+
+void mlx5_fpga_tls_send_teardown_cmd(struct mlx5_core_dev *mdev, void *flow,
+ u32 swid, gfp_t flags)
+{
+ struct mlx5_teardown_stream_context *ctx;
+ struct mlx5_fpga_dma_buf *buf;
+ void *cmd;
+
+ ctx = kzalloc(sizeof(*ctx) + MLX5_TLS_COMMAND_SIZE, flags);
+ if (!ctx)
+ return;
+
+ buf = &ctx->cmd.buf;
+ cmd = (ctx + 1);
+ MLX5_SET(tls_cmd, cmd, command_type, CMD_TEARDOWN_STREAM);
+ MLX5_SET(tls_cmd, cmd, swid, swid);
+
+ mlx5_fpga_tls_flow_to_cmd(flow, cmd);
+ kfree(flow);
+
+ buf->sg[0].data = cmd;
+ buf->sg[0].size = MLX5_TLS_COMMAND_SIZE;
+
+ ctx->swid = swid;
+ mlx5_fpga_tls_cmd_send(mdev->fpga, &ctx->cmd,
+ mlx5_fpga_tls_teardown_completion);
+}
+
+void mlx5_fpga_tls_del_tx_flow(struct mlx5_core_dev *mdev, u32 swid,
+ gfp_t flags)
+{
+ struct mlx5_fpga_tls *tls = mdev->fpga->tls;
+ void *flow;
+
+ rcu_read_lock();
+ flow = idr_find(&tls->tx_idr, swid);
+ rcu_read_unlock();
+
+ if (!flow) {
+ mlx5_fpga_err(mdev->fpga, "No flow information for swid %u\n",
+ swid);
+ return;
+ }
+
+ mlx5_fpga_tls_send_teardown_cmd(mdev, flow, swid, flags);
+}
+
+enum mlx5_fpga_setup_stream_status {
+ MLX5_FPGA_CMD_PENDING,
+ MLX5_FPGA_CMD_SEND_FAILED,
+ MLX5_FPGA_CMD_RESPONSE_RECEIVED,
+ MLX5_FPGA_CMD_ABANDONED,
+};
+
+struct mlx5_setup_stream_context {
+ struct mlx5_fpga_tls_command_context cmd;
+ atomic_t status;
+ u32 syndrome;
+ struct completion comp;
+};
+
+static void
+mlx5_fpga_tls_setup_completion(struct mlx5_fpga_conn *conn,
+ struct mlx5_fpga_device *fdev,
+ struct mlx5_fpga_tls_command_context *cmd,
+ struct mlx5_fpga_dma_buf *resp)
+{
+ struct mlx5_setup_stream_context *ctx =
+ container_of(cmd, struct mlx5_setup_stream_context, cmd);
+ int status = MLX5_FPGA_CMD_SEND_FAILED;
+ void *tls_cmd = ctx + 1;
+
+ /* If we failed to send to command resp == NULL */
+ if (resp) {
+ ctx->syndrome = MLX5_GET(tls_resp, resp->sg[0].data, syndrome);
+ status = MLX5_FPGA_CMD_RESPONSE_RECEIVED;
+ }
+
+ status = atomic_xchg_release(&ctx->status, status);
+ if (likely(status != MLX5_FPGA_CMD_ABANDONED)) {
+ complete(&ctx->comp);
+ return;
+ }
+
+ mlx5_fpga_err(fdev, "Command was abandoned, syndrome = %u\n",
+ ctx->syndrome);
+
+ if (!ctx->syndrome) {
+ /* The process was killed while waiting for the context to be
+ * added, and the add completed successfully.
+ * We need to destroy the HW context, and we can't can't reuse
+ * the command context because we might not have received
+ * the tx completion yet.
+ */
+ mlx5_fpga_tls_del_tx_flow(fdev->mdev,
+ MLX5_GET(tls_cmd, tls_cmd, swid),
+ GFP_ATOMIC);
+ }
+
+ mlx5_fpga_tls_put_command_ctx(cmd);
+}
+
+static int mlx5_fpga_tls_setup_stream_cmd(struct mlx5_core_dev *mdev,
+ struct mlx5_setup_stream_context *ctx)
+{
+ struct mlx5_fpga_dma_buf *buf;
+ void *cmd = ctx + 1;
+ int status, ret = 0;
+
+ buf = &ctx->cmd.buf;
+ buf->sg[0].data = cmd;
+ buf->sg[0].size = MLX5_TLS_COMMAND_SIZE;
+ MLX5_SET(tls_cmd, cmd, command_type, CMD_SETUP_STREAM);
+
+ init_completion(&ctx->comp);
+ atomic_set(&ctx->status, MLX5_FPGA_CMD_PENDING);
+ ctx->syndrome = -1;
+
+ mlx5_fpga_tls_cmd_send(mdev->fpga, &ctx->cmd,
+ mlx5_fpga_tls_setup_completion);
+ wait_for_completion_killable(&ctx->comp);
+
+ status = atomic_xchg_acquire(&ctx->status, MLX5_FPGA_CMD_ABANDONED);
+ if (unlikely(status == MLX5_FPGA_CMD_PENDING))
+ /* ctx is going to be released in mlx5_fpga_tls_setup_completion */
+ return -EINTR;
+
+ if (unlikely(ctx->syndrome))
+ ret = -ENOMEM;
+
+ mlx5_fpga_tls_put_command_ctx(&ctx->cmd);
+ return ret;
+}
+
+static void mlx5_fpga_tls_hw_qp_recv_cb(void *cb_arg,
+ struct mlx5_fpga_dma_buf *buf)
+{
+ struct mlx5_fpga_device *fdev = (struct mlx5_fpga_device *)cb_arg;
+
+ mlx5_fpga_tls_cmd_complete(fdev, buf);
+}
+
+bool mlx5_fpga_is_tls_device(struct mlx5_core_dev *mdev)
+{
+ if (!mdev->fpga || !MLX5_CAP_GEN(mdev, fpga))
+ return false;
+
+ if (MLX5_CAP_FPGA(mdev, ieee_vendor_id) !=
+ MLX5_FPGA_CAP_SANDBOX_VENDOR_ID_MLNX)
+ return false;
+
+ if (MLX5_CAP_FPGA(mdev, sandbox_product_id) !=
+ MLX5_FPGA_CAP_SANDBOX_PRODUCT_ID_TLS)
+ return false;
+
+ if (MLX5_CAP_FPGA(mdev, sandbox_product_version) != 0)
+ return false;
+
+ return true;
+}
+
+static int mlx5_fpga_tls_get_caps(struct mlx5_fpga_device *fdev,
+ u32 *p_caps)
+{
+ int err, cap_size = MLX5_ST_SZ_BYTES(tls_extended_cap);
+ u32 caps = 0;
+ void *buf;
+
+ buf = kzalloc(cap_size, GFP_KERNEL);
+ if (!buf)
+ return -ENOMEM;
+
+ err = mlx5_fpga_get_sbu_caps(fdev, cap_size, buf);
+ if (err)
+ goto out;
+
+ if (MLX5_GET(tls_extended_cap, buf, tx))
+ caps |= MLX5_ACCEL_TLS_TX;
+ if (MLX5_GET(tls_extended_cap, buf, rx))
+ caps |= MLX5_ACCEL_TLS_RX;
+ if (MLX5_GET(tls_extended_cap, buf, tls_v12))
+ caps |= MLX5_ACCEL_TLS_V12;
+ if (MLX5_GET(tls_extended_cap, buf, tls_v13))
+ caps |= MLX5_ACCEL_TLS_V13;
+ if (MLX5_GET(tls_extended_cap, buf, lro))
+ caps |= MLX5_ACCEL_TLS_LRO;
+ if (MLX5_GET(tls_extended_cap, buf, ipv6))
+ caps |= MLX5_ACCEL_TLS_IPV6;
+
+ if (MLX5_GET(tls_extended_cap, buf, aes_gcm_128))
+ caps |= MLX5_ACCEL_TLS_AES_GCM128;
+ if (MLX5_GET(tls_extended_cap, buf, aes_gcm_256))
+ caps |= MLX5_ACCEL_TLS_AES_GCM256;
+
+ *p_caps = caps;
+ err = 0;
+out:
+ kfree(buf);
+ return err;
+}
+
+int mlx5_fpga_tls_init(struct mlx5_core_dev *mdev)
+{
+ struct mlx5_fpga_device *fdev = mdev->fpga;
+ struct mlx5_fpga_conn_attr init_attr = {0};
+ struct mlx5_fpga_conn *conn;
+ struct mlx5_fpga_tls *tls;
+ int err = 0;
+
+ if (!mlx5_fpga_is_tls_device(mdev) || !fdev)
+ return 0;
+
+ tls = kzalloc(sizeof(*tls), GFP_KERNEL);
+ if (!tls)
+ return -ENOMEM;
+
+ err = mlx5_fpga_tls_get_caps(fdev, &tls->caps);
+ if (err)
+ goto error;
+
+ if (!(tls->caps & (MLX5_ACCEL_TLS_TX | MLX5_ACCEL_TLS_V12 |
+ MLX5_ACCEL_TLS_AES_GCM128))) {
+ err = -ENOTSUPP;
+ goto error;
+ }
+
+ init_attr.rx_size = SBU_QP_QUEUE_SIZE;
+ init_attr.tx_size = SBU_QP_QUEUE_SIZE;
+ init_attr.recv_cb = mlx5_fpga_tls_hw_qp_recv_cb;
+ init_attr.cb_arg = fdev;
+ conn = mlx5_fpga_sbu_conn_create(fdev, &init_attr);
+ if (IS_ERR(conn)) {
+ err = PTR_ERR(conn);
+ mlx5_fpga_err(fdev, "Error creating TLS command connection %d\n",
+ err);
+ goto error;
+ }
+
+ tls->conn = conn;
+ spin_lock_init(&tls->pending_cmds_lock);
+ INIT_LIST_HEAD(&tls->pending_cmds);
+
+ idr_init(&tls->tx_idr);
+ spin_lock_init(&tls->idr_spinlock);
+ fdev->tls = tls;
+ return 0;
+
+error:
+ kfree(tls);
+ return err;
+}
+
+void mlx5_fpga_tls_cleanup(struct mlx5_core_dev *mdev)
+{
+ struct mlx5_fpga_device *fdev = mdev->fpga;
+
+ if (!fdev || !fdev->tls)
+ return;
+
+ mlx5_fpga_sbu_conn_destroy(fdev->tls->conn);
+ kfree(fdev->tls);
+ fdev->tls = NULL;
+}
+
+static void mlx5_fpga_tls_set_aes_gcm128_ctx(void *cmd,
+ struct tls_crypto_info *info,
+ __be64 *rcd_sn)
+{
+ struct tls12_crypto_info_aes_gcm_128 *crypto_info =
+ (struct tls12_crypto_info_aes_gcm_128 *)info;
+
+ memcpy(MLX5_ADDR_OF(tls_cmd, cmd, tls_rcd_sn), crypto_info->rec_seq,
+ TLS_CIPHER_AES_GCM_128_REC_SEQ_SIZE);
+
+ memcpy(MLX5_ADDR_OF(tls_cmd, cmd, tls_implicit_iv),
+ crypto_info->salt, TLS_CIPHER_AES_GCM_128_SALT_SIZE);
+ memcpy(MLX5_ADDR_OF(tls_cmd, cmd, encryption_key),
+ crypto_info->key, TLS_CIPHER_AES_GCM_128_KEY_SIZE);
+
+ /* in AES-GCM 128 we need to write the key twice */
+ memcpy(MLX5_ADDR_OF(tls_cmd, cmd, encryption_key) +
+ TLS_CIPHER_AES_GCM_128_KEY_SIZE,
+ crypto_info->key, TLS_CIPHER_AES_GCM_128_KEY_SIZE);
+
+ MLX5_SET(tls_cmd, cmd, alg, MLX5_TLS_ALG_AES_GCM_128);
+}
+
+static int mlx5_fpga_tls_set_key_material(void *cmd, u32 caps,
+ struct tls_crypto_info *crypto_info)
+{
+ __be64 rcd_sn;
+
+ switch (crypto_info->cipher_type) {
+ case TLS_CIPHER_AES_GCM_128:
+ if (!(caps & MLX5_ACCEL_TLS_AES_GCM128))
+ return -EINVAL;
+ mlx5_fpga_tls_set_aes_gcm128_ctx(cmd, crypto_info, &rcd_sn);
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int mlx5_fpga_tls_add_flow(struct mlx5_core_dev *mdev, void *flow,
+ struct tls_crypto_info *crypto_info, u32 swid,
+ u32 tcp_sn)
+{
+ u32 caps = mlx5_fpga_tls_device_caps(mdev);
+ struct mlx5_setup_stream_context *ctx;
+ int ret = -ENOMEM;
+ size_t cmd_size;
+ void *cmd;
+
+ cmd_size = MLX5_TLS_COMMAND_SIZE + sizeof(*ctx);
+ ctx = kzalloc(cmd_size, GFP_KERNEL);
+ if (!ctx)
+ goto out;
+
+ cmd = ctx + 1;
+ ret = mlx5_fpga_tls_set_key_material(cmd, caps, crypto_info);
+ if (ret)
+ goto free_ctx;
+
+ mlx5_fpga_tls_flow_to_cmd(flow, cmd);
+
+ MLX5_SET(tls_cmd, cmd, swid, swid);
+ MLX5_SET(tls_cmd, cmd, tcp_sn, tcp_sn);
+
+ return mlx5_fpga_tls_setup_stream_cmd(mdev, ctx);
+
+free_ctx:
+ kfree(ctx);
+out:
+ return ret;
+}
+
+int mlx5_fpga_tls_add_tx_flow(struct mlx5_core_dev *mdev, void *flow,
+ struct tls_crypto_info *crypto_info,
+ u32 start_offload_tcp_sn, u32 *p_swid)
+{
+ struct mlx5_fpga_tls *tls = mdev->fpga->tls;
+ int ret = -ENOMEM;
+ u32 swid;
+
+ ret = mlx5_fpga_tls_alloc_swid(&tls->tx_idr, &tls->idr_spinlock, flow);
+ if (ret < 0)
+ return ret;
+
+ swid = ret;
+ MLX5_SET(tls_flow, flow, direction_sx, 1);
+
+ ret = mlx5_fpga_tls_add_flow(mdev, flow, crypto_info, swid,
+ start_offload_tcp_sn);
+ if (ret && ret != -EINTR)
+ goto free_swid;
+
+ *p_swid = swid;
+ return 0;
+free_swid:
+ mlx5_fpga_tls_release_swid(&tls->tx_idr, &tls->idr_spinlock, swid);
+
+ return ret;
+}
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h
new file mode 100644
index 0000000..800a214
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.h
@@ -0,0 +1,68 @@
+/*
+ * Copyright (c) 2018 Mellanox Technologies. All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses. You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ * Redistribution and use in source and binary forms, with or
+ * without modification, are permitted provided that the following
+ * conditions are met:
+ *
+ * - Redistributions of source code must retain the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer.
+ *
+ * - Redistributions in binary form must reproduce the above
+ * copyright notice, this list of conditions and the following
+ * disclaimer in the documentation and/or other materials
+ * provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ *
+ */
+
+#ifndef __MLX5_FPGA_TLS_H__
+#define __MLX5_FPGA_TLS_H__
+
+#include <linux/mlx5/driver.h>
+
+#include <net/tls.h>
+#include "fpga/core.h"
+
+struct mlx5_fpga_tls {
+ struct list_head pending_cmds;
+ spinlock_t pending_cmds_lock; /* Protects pending_cmds */
+ u32 caps;
+ struct mlx5_fpga_conn *conn;
+
+ struct idr tx_idr;
+ spinlock_t idr_spinlock; /* protects the IDR */
+};
+
+int mlx5_fpga_tls_add_tx_flow(struct mlx5_core_dev *mdev, void *flow,
+ struct tls_crypto_info *crypto_info,
+ u32 start_offload_tcp_sn, u32 *p_swid);
+
+void mlx5_fpga_tls_del_tx_flow(struct mlx5_core_dev *mdev, u32 swid,
+ gfp_t flags);
+
+bool mlx5_fpga_is_tls_device(struct mlx5_core_dev *mdev);
+int mlx5_fpga_tls_init(struct mlx5_core_dev *mdev);
+void mlx5_fpga_tls_cleanup(struct mlx5_core_dev *mdev);
+
+static inline u32 mlx5_fpga_tls_device_caps(struct mlx5_core_dev *mdev)
+{
+ return mdev->fpga->tls->caps;
+}
+
+#endif /* __MLX5_FPGA_TLS_H__ */
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 63a8ea3..b865597 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -60,6 +60,7 @@
#include "fpga/core.h"
#include "fpga/ipsec.h"
#include "accel/ipsec.h"
+#include "accel/tls.h"
#include "lib/clock.h"
MODULE_AUTHOR("Eli Cohen <eli@mellanox.com>");
@@ -1190,6 +1191,12 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
goto err_ipsec_start;
}
+ err = mlx5_accel_tls_init(dev);
+ if (err) {
+ dev_err(&pdev->dev, "TLS device start failed %d\n", err);
+ goto err_tls_start;
+ }
+
err = mlx5_init_fs(dev);
if (err) {
dev_err(&pdev->dev, "Failed to init flow steering\n");
@@ -1231,6 +1238,9 @@ static int mlx5_load_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
mlx5_cleanup_fs(dev);
err_fs:
+ mlx5_accel_tls_cleanup(dev);
+
+err_tls_start:
mlx5_accel_ipsec_cleanup(dev);
err_ipsec_start:
@@ -1306,6 +1316,7 @@ static int mlx5_unload_one(struct mlx5_core_dev *dev, struct mlx5_priv *priv,
mlx5_sriov_detach(dev);
mlx5_cleanup_fs(dev);
mlx5_accel_ipsec_cleanup(dev);
+ mlx5_accel_tls_cleanup(dev);
mlx5_fpga_device_stop(dev);
mlx5_irq_clear_affinity_hints(dev);
free_comp_eqs(dev);
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 1aad455..b8918a1 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -356,22 +356,6 @@ struct mlx5_ifc_odp_per_transport_service_cap_bits {
u8 reserved_at_6[0x1a];
};
-struct mlx5_ifc_ipv4_layout_bits {
- u8 reserved_at_0[0x60];
-
- u8 ipv4[0x20];
-};
-
-struct mlx5_ifc_ipv6_layout_bits {
- u8 ipv6[16][0x8];
-};
-
-union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits {
- struct mlx5_ifc_ipv6_layout_bits ipv6_layout;
- struct mlx5_ifc_ipv4_layout_bits ipv4_layout;
- u8 reserved_at_0[0x80];
-};
-
struct mlx5_ifc_fte_match_set_lyr_2_4_bits {
u8 smac_47_16[0x20];
diff --git a/include/linux/mlx5/mlx5_ifc_fpga.h b/include/linux/mlx5/mlx5_ifc_fpga.h
index ec05249..1930915 100644
--- a/include/linux/mlx5/mlx5_ifc_fpga.h
+++ b/include/linux/mlx5/mlx5_ifc_fpga.h
@@ -32,12 +32,29 @@
#ifndef MLX5_IFC_FPGA_H
#define MLX5_IFC_FPGA_H
+struct mlx5_ifc_ipv4_layout_bits {
+ u8 reserved_at_0[0x60];
+
+ u8 ipv4[0x20];
+};
+
+struct mlx5_ifc_ipv6_layout_bits {
+ u8 ipv6[16][0x8];
+};
+
+union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits {
+ struct mlx5_ifc_ipv6_layout_bits ipv6_layout;
+ struct mlx5_ifc_ipv4_layout_bits ipv4_layout;
+ u8 reserved_at_0[0x80];
+};
+
enum {
MLX5_FPGA_CAP_SANDBOX_VENDOR_ID_MLNX = 0x2c9,
};
enum {
MLX5_FPGA_CAP_SANDBOX_PRODUCT_ID_IPSEC = 0x2,
+ MLX5_FPGA_CAP_SANDBOX_PRODUCT_ID_TLS = 0x3,
};
struct mlx5_ifc_fpga_shell_caps_bits {
@@ -370,6 +387,27 @@ struct mlx5_ifc_fpga_destroy_qp_out_bits {
u8 reserved_at_40[0x40];
};
+struct mlx5_ifc_tls_extended_cap_bits {
+ u8 aes_gcm_128[0x1];
+ u8 aes_gcm_256[0x1];
+ u8 reserved_at_2[0x1e];
+ u8 reserved_at_20[0x20];
+ u8 context_capacity_total[0x20];
+ u8 context_capacity_rx[0x20];
+ u8 context_capacity_tx[0x20];
+ u8 reserved_at_a0[0x10];
+ u8 tls_counter_size[0x10];
+ u8 tls_counters_addr_low[0x20];
+ u8 tls_counters_addr_high[0x20];
+ u8 rx[0x1];
+ u8 tx[0x1];
+ u8 tls_v12[0x1];
+ u8 tls_v13[0x1];
+ u8 lro[0x1];
+ u8 ipv6[0x1];
+ u8 reserved_at_106[0x1a];
+};
+
struct mlx5_ifc_ipsec_extended_cap_bits {
u8 encapsulation[0x20];
@@ -519,4 +557,43 @@ struct mlx5_ifc_fpga_ipsec_sa {
__be16 reserved2;
} __packed;
+enum fpga_tls_cmds {
+ CMD_SETUP_STREAM = 0x1001,
+ CMD_TEARDOWN_STREAM = 0x1002,
+};
+
+#define MLX5_TLS_1_2 (0)
+
+#define MLX5_TLS_ALG_AES_GCM_128 (0)
+#define MLX5_TLS_ALG_AES_GCM_256 (1)
+
+struct mlx5_ifc_tls_cmd_bits {
+ u8 command_type[0x20];
+ u8 ipv6[0x1];
+ u8 direction_sx[0x1];
+ u8 tls_version[0x2];
+ u8 reserved[0x1c];
+ u8 swid[0x20];
+ u8 src_port[0x10];
+ u8 dst_port[0x10];
+ union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits src_ipv4_src_ipv6;
+ union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits dst_ipv4_dst_ipv6;
+ u8 tls_rcd_sn[0x40];
+ u8 tcp_sn[0x20];
+ u8 tls_implicit_iv[0x20];
+ u8 tls_xor_iv[0x40];
+ u8 encryption_key[0x100];
+ u8 alg[4];
+ u8 reserved2[0x1c];
+ u8 reserved3[0x4a0];
+};
+
+struct mlx5_ifc_tls_resp_bits {
+ u8 syndrome[0x20];
+ u8 stream_id[0x20];
+ u8 reserverd[0x40];
+};
+
+#define MLX5_TLS_COMMAND_SIZE (0x100)
+
#endif /* MLX5_IFC_FPGA_H */
--
1.8.3.1
^ permalink raw reply related
* [PATCH V10 net-next 06/14] net/tls: Split conf to rx + tx
From: Boris Pismenny @ 2018-04-30 7:16 UTC (permalink / raw)
To: davem; +Cc: netdev, saeedm, borisp, davejwatson, ktkhai, sergei.shtylyov
In-Reply-To: <1525072583-138506-1-git-send-email-borisp@mellanox.com>
In TLS inline crypto, we can have one direction in software
and another in hardware. Thus, we split the TLS configuration to separate
structures for receive and transmit.
Signed-off-by: Boris Pismenny <borisp@mellanox.com>
---
include/net/tls.h | 51 +++++++++++++-------
net/tls/tls_main.c | 103 ++++++++++++++++++++-------------------
net/tls/tls_sw.c | 138 ++++++++++++++++++++++++++++++-----------------------
3 files changed, 163 insertions(+), 129 deletions(-)
diff --git a/include/net/tls.h b/include/net/tls.h
index 3da8e13..95a8c60 100644
--- a/include/net/tls.h
+++ b/include/net/tls.h
@@ -83,21 +83,10 @@ struct tls_device {
void (*unhash)(struct tls_device *device, struct sock *sk);
};
-struct tls_sw_context {
+struct tls_sw_context_tx {
struct crypto_aead *aead_send;
- struct crypto_aead *aead_recv;
struct crypto_wait async_wait;
- /* Receive context */
- struct strparser strp;
- void (*saved_data_ready)(struct sock *sk);
- unsigned int (*sk_poll)(struct file *file, struct socket *sock,
- struct poll_table_struct *wait);
- struct sk_buff *recv_pkt;
- u8 control;
- bool decrypted;
-
- /* Sending context */
char aad_space[TLS_AAD_SPACE_SIZE];
unsigned int sg_plaintext_size;
@@ -114,6 +103,19 @@ struct tls_sw_context {
struct scatterlist sg_aead_out[2];
};
+struct tls_sw_context_rx {
+ struct crypto_aead *aead_recv;
+ struct crypto_wait async_wait;
+
+ struct strparser strp;
+ void (*saved_data_ready)(struct sock *sk);
+ unsigned int (*sk_poll)(struct file *file, struct socket *sock,
+ struct poll_table_struct *wait);
+ struct sk_buff *recv_pkt;
+ u8 control;
+ bool decrypted;
+};
+
enum {
TLS_PENDING_CLOSED_RECORD
};
@@ -138,9 +140,15 @@ struct tls_context {
struct tls12_crypto_info_aes_gcm_128 crypto_recv_aes_gcm_128;
};
- void *priv_ctx;
+ struct list_head list;
+ struct net_device *netdev;
+ refcount_t refcount;
+
+ void *priv_ctx_tx;
+ void *priv_ctx_rx;
- u8 conf:3;
+ u8 tx_conf:3;
+ u8 rx_conf:3;
struct cipher_context tx;
struct cipher_context rx;
@@ -177,7 +185,8 @@ int tls_sk_attach(struct sock *sk, int optname, char __user *optval,
int tls_sw_sendpage(struct sock *sk, struct page *page,
int offset, size_t size, int flags);
void tls_sw_close(struct sock *sk, long timeout);
-void tls_sw_free_resources(struct sock *sk);
+void tls_sw_free_resources_tx(struct sock *sk);
+void tls_sw_free_resources_rx(struct sock *sk);
int tls_sw_recvmsg(struct sock *sk, struct msghdr *msg, size_t len,
int nonblock, int flags, int *addr_len);
unsigned int tls_sw_poll(struct file *file, struct socket *sock,
@@ -297,16 +306,22 @@ static inline struct tls_context *tls_get_ctx(const struct sock *sk)
return icsk->icsk_ulp_data;
}
-static inline struct tls_sw_context *tls_sw_ctx(
+static inline struct tls_sw_context_rx *tls_sw_ctx_rx(
+ const struct tls_context *tls_ctx)
+{
+ return (struct tls_sw_context_rx *)tls_ctx->priv_ctx_rx;
+}
+
+static inline struct tls_sw_context_tx *tls_sw_ctx_tx(
const struct tls_context *tls_ctx)
{
- return (struct tls_sw_context *)tls_ctx->priv_ctx;
+ return (struct tls_sw_context_tx *)tls_ctx->priv_ctx_tx;
}
static inline struct tls_offload_context *tls_offload_ctx(
const struct tls_context *tls_ctx)
{
- return (struct tls_offload_context *)tls_ctx->priv_ctx;
+ return (struct tls_offload_context *)tls_ctx->priv_ctx_tx;
}
int tls_proccess_cmsg(struct sock *sk, struct msghdr *msg,
diff --git a/net/tls/tls_main.c b/net/tls/tls_main.c
index 0d37997..545bf34 100644
--- a/net/tls/tls_main.c
+++ b/net/tls/tls_main.c
@@ -51,12 +51,9 @@ enum {
TLSV6,
TLS_NUM_PROTS,
};
-
enum {
TLS_BASE,
- TLS_SW_TX,
- TLS_SW_RX,
- TLS_SW_RXTX,
+ TLS_SW,
TLS_HW_RECORD,
TLS_NUM_CONFIG,
};
@@ -65,14 +62,14 @@ enum {
static DEFINE_MUTEX(tcpv6_prot_mutex);
static LIST_HEAD(device_list);
static DEFINE_MUTEX(device_mutex);
-static struct proto tls_prots[TLS_NUM_PROTS][TLS_NUM_CONFIG];
+static struct proto tls_prots[TLS_NUM_PROTS][TLS_NUM_CONFIG][TLS_NUM_CONFIG];
static struct proto_ops tls_sw_proto_ops;
-static inline void update_sk_prot(struct sock *sk, struct tls_context *ctx)
+static void update_sk_prot(struct sock *sk, struct tls_context *ctx)
{
int ip_ver = sk->sk_family == AF_INET6 ? TLSV6 : TLSV4;
- sk->sk_prot = &tls_prots[ip_ver][ctx->conf];
+ sk->sk_prot = &tls_prots[ip_ver][ctx->tx_conf][ctx->rx_conf];
}
int wait_on_pending_writer(struct sock *sk, long *timeo)
@@ -245,10 +242,10 @@ static void tls_sk_proto_close(struct sock *sk, long timeout)
lock_sock(sk);
sk_proto_close = ctx->sk_proto_close;
- if (ctx->conf == TLS_HW_RECORD)
+ if (ctx->tx_conf == TLS_HW_RECORD && ctx->rx_conf == TLS_HW_RECORD)
goto skip_tx_cleanup;
- if (ctx->conf == TLS_BASE) {
+ if (ctx->tx_conf == TLS_BASE && ctx->rx_conf == TLS_BASE) {
kfree(ctx);
ctx = NULL;
goto skip_tx_cleanup;
@@ -270,15 +267,17 @@ static void tls_sk_proto_close(struct sock *sk, long timeout)
}
}
- kfree(ctx->tx.rec_seq);
- kfree(ctx->tx.iv);
- kfree(ctx->rx.rec_seq);
- kfree(ctx->rx.iv);
+ /* We need these for tls_sw_fallback handling of other packets */
+ if (ctx->tx_conf == TLS_SW) {
+ kfree(ctx->tx.rec_seq);
+ kfree(ctx->tx.iv);
+ tls_sw_free_resources_tx(sk);
+ }
- if (ctx->conf == TLS_SW_TX ||
- ctx->conf == TLS_SW_RX ||
- ctx->conf == TLS_SW_RXTX) {
- tls_sw_free_resources(sk);
+ if (ctx->rx_conf == TLS_SW) {
+ kfree(ctx->rx.rec_seq);
+ kfree(ctx->rx.iv);
+ tls_sw_free_resources_rx(sk);
}
skip_tx_cleanup:
@@ -287,7 +286,8 @@ static void tls_sk_proto_close(struct sock *sk, long timeout)
/* free ctx for TLS_HW_RECORD, used by tcp_set_state
* for sk->sk_prot->unhash [tls_hw_unhash]
*/
- if (ctx && ctx->conf == TLS_HW_RECORD)
+ if (ctx && ctx->tx_conf == TLS_HW_RECORD &&
+ ctx->rx_conf == TLS_HW_RECORD)
kfree(ctx);
}
@@ -441,25 +441,21 @@ static int do_tls_setsockopt_conf(struct sock *sk, char __user *optval,
goto err_crypto_info;
}
- /* currently SW is default, we will have ethtool in future */
if (tx) {
rc = tls_set_sw_offload(sk, ctx, 1);
- if (ctx->conf == TLS_SW_RX)
- conf = TLS_SW_RXTX;
- else
- conf = TLS_SW_TX;
+ conf = TLS_SW;
} else {
rc = tls_set_sw_offload(sk, ctx, 0);
- if (ctx->conf == TLS_SW_TX)
- conf = TLS_SW_RXTX;
- else
- conf = TLS_SW_RX;
+ conf = TLS_SW;
}
if (rc)
goto err_crypto_info;
- ctx->conf = conf;
+ if (tx)
+ ctx->tx_conf = conf;
+ else
+ ctx->rx_conf = conf;
update_sk_prot(sk, ctx);
if (tx) {
ctx->sk_write_space = sk->sk_write_space;
@@ -535,7 +531,8 @@ static int tls_hw_prot(struct sock *sk)
ctx->hash = sk->sk_prot->hash;
ctx->unhash = sk->sk_prot->unhash;
ctx->sk_proto_close = sk->sk_prot->close;
- ctx->conf = TLS_HW_RECORD;
+ ctx->rx_conf = TLS_HW_RECORD;
+ ctx->tx_conf = TLS_HW_RECORD;
update_sk_prot(sk, ctx);
rc = 1;
break;
@@ -579,29 +576,30 @@ static int tls_hw_hash(struct sock *sk)
return err;
}
-static void build_protos(struct proto *prot, struct proto *base)
+static void build_protos(struct proto prot[TLS_NUM_CONFIG][TLS_NUM_CONFIG],
+ struct proto *base)
{
- prot[TLS_BASE] = *base;
- prot[TLS_BASE].setsockopt = tls_setsockopt;
- prot[TLS_BASE].getsockopt = tls_getsockopt;
- prot[TLS_BASE].close = tls_sk_proto_close;
-
- prot[TLS_SW_TX] = prot[TLS_BASE];
- prot[TLS_SW_TX].sendmsg = tls_sw_sendmsg;
- prot[TLS_SW_TX].sendpage = tls_sw_sendpage;
-
- prot[TLS_SW_RX] = prot[TLS_BASE];
- prot[TLS_SW_RX].recvmsg = tls_sw_recvmsg;
- prot[TLS_SW_RX].close = tls_sk_proto_close;
-
- prot[TLS_SW_RXTX] = prot[TLS_SW_TX];
- prot[TLS_SW_RXTX].recvmsg = tls_sw_recvmsg;
- prot[TLS_SW_RXTX].close = tls_sk_proto_close;
-
- prot[TLS_HW_RECORD] = *base;
- prot[TLS_HW_RECORD].hash = tls_hw_hash;
- prot[TLS_HW_RECORD].unhash = tls_hw_unhash;
- prot[TLS_HW_RECORD].close = tls_sk_proto_close;
+ prot[TLS_BASE][TLS_BASE] = *base;
+ prot[TLS_BASE][TLS_BASE].setsockopt = tls_setsockopt;
+ prot[TLS_BASE][TLS_BASE].getsockopt = tls_getsockopt;
+ prot[TLS_BASE][TLS_BASE].close = tls_sk_proto_close;
+
+ prot[TLS_SW][TLS_BASE] = prot[TLS_BASE][TLS_BASE];
+ prot[TLS_SW][TLS_BASE].sendmsg = tls_sw_sendmsg;
+ prot[TLS_SW][TLS_BASE].sendpage = tls_sw_sendpage;
+
+ prot[TLS_BASE][TLS_SW] = prot[TLS_BASE][TLS_BASE];
+ prot[TLS_BASE][TLS_SW].recvmsg = tls_sw_recvmsg;
+ prot[TLS_BASE][TLS_SW].close = tls_sk_proto_close;
+
+ prot[TLS_SW][TLS_SW] = prot[TLS_SW][TLS_BASE];
+ prot[TLS_SW][TLS_SW].recvmsg = tls_sw_recvmsg;
+ prot[TLS_SW][TLS_SW].close = tls_sk_proto_close;
+
+ prot[TLS_HW_RECORD][TLS_HW_RECORD] = *base;
+ prot[TLS_HW_RECORD][TLS_HW_RECORD].hash = tls_hw_hash;
+ prot[TLS_HW_RECORD][TLS_HW_RECORD].unhash = tls_hw_unhash;
+ prot[TLS_HW_RECORD][TLS_HW_RECORD].close = tls_sk_proto_close;
}
static int tls_init(struct sock *sk)
@@ -643,7 +641,8 @@ static int tls_init(struct sock *sk)
mutex_unlock(&tcpv6_prot_mutex);
}
- ctx->conf = TLS_BASE;
+ ctx->tx_conf = TLS_BASE;
+ ctx->rx_conf = TLS_BASE;
update_sk_prot(sk, ctx);
out:
return rc;
diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
index 6ed1c02..5c3909c 100644
--- a/net/tls/tls_sw.c
+++ b/net/tls/tls_sw.c
@@ -52,7 +52,7 @@ static int tls_do_decryption(struct sock *sk,
gfp_t flags)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
struct strp_msg *rxm = strp_msg(skb);
struct aead_request *aead_req;
@@ -122,7 +122,7 @@ static void trim_sg(struct sock *sk, struct scatterlist *sg,
static void trim_both_sgl(struct sock *sk, int target_size)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
trim_sg(sk, ctx->sg_plaintext_data,
&ctx->sg_plaintext_num_elem,
@@ -141,7 +141,7 @@ static void trim_both_sgl(struct sock *sk, int target_size)
static int alloc_encrypted_sg(struct sock *sk, int len)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
int rc = 0;
rc = sk_alloc_sg(sk, len,
@@ -155,7 +155,7 @@ static int alloc_encrypted_sg(struct sock *sk, int len)
static int alloc_plaintext_sg(struct sock *sk, int len)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
int rc = 0;
rc = sk_alloc_sg(sk, len, ctx->sg_plaintext_data, 0,
@@ -181,7 +181,7 @@ static void free_sg(struct sock *sk, struct scatterlist *sg,
static void tls_free_both_sg(struct sock *sk)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
free_sg(sk, ctx->sg_encrypted_data, &ctx->sg_encrypted_num_elem,
&ctx->sg_encrypted_size);
@@ -191,7 +191,7 @@ static void tls_free_both_sg(struct sock *sk)
}
static int tls_do_encryption(struct tls_context *tls_ctx,
- struct tls_sw_context *ctx, size_t data_len,
+ struct tls_sw_context_tx *ctx, size_t data_len,
gfp_t flags)
{
unsigned int req_size = sizeof(struct aead_request) +
@@ -227,7 +227,7 @@ static int tls_push_record(struct sock *sk, int flags,
unsigned char record_type)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
int rc;
sg_mark_end(ctx->sg_plaintext_data + ctx->sg_plaintext_num_elem - 1);
@@ -339,7 +339,7 @@ static int memcopy_from_iter(struct sock *sk, struct iov_iter *from,
int bytes)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
struct scatterlist *sg = ctx->sg_plaintext_data;
int copy, i, rc = 0;
@@ -367,7 +367,7 @@ static int memcopy_from_iter(struct sock *sk, struct iov_iter *from,
int tls_sw_sendmsg(struct sock *sk, struct msghdr *msg, size_t size)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
int ret = 0;
int required_size;
long timeo = sock_sndtimeo(sk, msg->msg_flags & MSG_DONTWAIT);
@@ -522,7 +522,7 @@ int tls_sw_sendpage(struct sock *sk, struct page *page,
int offset, size_t size, int flags)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
int ret = 0;
long timeo = sock_sndtimeo(sk, flags & MSG_DONTWAIT);
bool eor;
@@ -636,7 +636,7 @@ static struct sk_buff *tls_wait_data(struct sock *sk, int flags,
long timeo, int *err)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
struct sk_buff *skb;
DEFINE_WAIT_FUNC(wait, woken_wake_function);
@@ -674,7 +674,7 @@ static int decrypt_skb(struct sock *sk, struct sk_buff *skb,
struct scatterlist *sgout)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
char iv[TLS_CIPHER_AES_GCM_128_SALT_SIZE + MAX_IV_SIZE];
struct scatterlist sgin_arr[MAX_SKB_FRAGS + 2];
struct scatterlist *sgin = &sgin_arr[0];
@@ -723,7 +723,7 @@ static bool tls_sw_advance_skb(struct sock *sk, struct sk_buff *skb,
unsigned int len)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
struct strp_msg *rxm = strp_msg(skb);
if (len < rxm->full_len) {
@@ -749,7 +749,7 @@ int tls_sw_recvmsg(struct sock *sk,
int *addr_len)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
unsigned char control;
struct strp_msg *rxm;
struct sk_buff *skb;
@@ -869,7 +869,7 @@ ssize_t tls_sw_splice_read(struct socket *sock, loff_t *ppos,
size_t len, unsigned int flags)
{
struct tls_context *tls_ctx = tls_get_ctx(sock->sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
struct strp_msg *rxm = NULL;
struct sock *sk = sock->sk;
struct sk_buff *skb;
@@ -922,7 +922,7 @@ unsigned int tls_sw_poll(struct file *file, struct socket *sock,
unsigned int ret;
struct sock *sk = sock->sk;
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
/* Grab POLLOUT and POLLHUP from the underlying socket */
ret = ctx->sk_poll(file, sock, wait);
@@ -938,7 +938,7 @@ unsigned int tls_sw_poll(struct file *file, struct socket *sock,
static int tls_read_size(struct strparser *strp, struct sk_buff *skb)
{
struct tls_context *tls_ctx = tls_get_ctx(strp->sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
char header[tls_ctx->rx.prepend_size];
struct strp_msg *rxm = strp_msg(skb);
size_t cipher_overhead;
@@ -987,7 +987,7 @@ static int tls_read_size(struct strparser *strp, struct sk_buff *skb)
static void tls_queue(struct strparser *strp, struct sk_buff *skb)
{
struct tls_context *tls_ctx = tls_get_ctx(strp->sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
struct strp_msg *rxm;
rxm = strp_msg(skb);
@@ -1003,18 +1003,28 @@ static void tls_queue(struct strparser *strp, struct sk_buff *skb)
static void tls_data_ready(struct sock *sk)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
strp_data_ready(&ctx->strp);
}
-void tls_sw_free_resources(struct sock *sk)
+void tls_sw_free_resources_tx(struct sock *sk)
{
struct tls_context *tls_ctx = tls_get_ctx(sk);
- struct tls_sw_context *ctx = tls_sw_ctx(tls_ctx);
+ struct tls_sw_context_tx *ctx = tls_sw_ctx_tx(tls_ctx);
if (ctx->aead_send)
crypto_free_aead(ctx->aead_send);
+ tls_free_both_sg(sk);
+
+ kfree(ctx);
+}
+
+void tls_sw_free_resources_rx(struct sock *sk)
+{
+ struct tls_context *tls_ctx = tls_get_ctx(sk);
+ struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
+
if (ctx->aead_recv) {
if (ctx->recv_pkt) {
kfree_skb(ctx->recv_pkt);
@@ -1030,10 +1040,7 @@ void tls_sw_free_resources(struct sock *sk)
lock_sock(sk);
}
- tls_free_both_sg(sk);
-
kfree(ctx);
- kfree(tls_ctx);
}
int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx)
@@ -1041,7 +1048,8 @@ int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx)
char keyval[TLS_CIPHER_AES_GCM_128_KEY_SIZE];
struct tls_crypto_info *crypto_info;
struct tls12_crypto_info_aes_gcm_128 *gcm_128_info;
- struct tls_sw_context *sw_ctx;
+ struct tls_sw_context_tx *sw_ctx_tx = NULL;
+ struct tls_sw_context_rx *sw_ctx_rx = NULL;
struct cipher_context *cctx;
struct crypto_aead **aead;
struct strp_callbacks cb;
@@ -1054,27 +1062,32 @@ int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx)
goto out;
}
- if (!ctx->priv_ctx) {
- sw_ctx = kzalloc(sizeof(*sw_ctx), GFP_KERNEL);
- if (!sw_ctx) {
+ if (tx) {
+ sw_ctx_tx = kzalloc(sizeof(*sw_ctx_tx), GFP_KERNEL);
+ if (!sw_ctx_tx) {
rc = -ENOMEM;
goto out;
}
- crypto_init_wait(&sw_ctx->async_wait);
+ crypto_init_wait(&sw_ctx_tx->async_wait);
+ ctx->priv_ctx_tx = sw_ctx_tx;
} else {
- sw_ctx = ctx->priv_ctx;
+ sw_ctx_rx = kzalloc(sizeof(*sw_ctx_rx), GFP_KERNEL);
+ if (!sw_ctx_rx) {
+ rc = -ENOMEM;
+ goto out;
+ }
+ crypto_init_wait(&sw_ctx_rx->async_wait);
+ ctx->priv_ctx_rx = sw_ctx_rx;
}
- ctx->priv_ctx = (struct tls_offload_context *)sw_ctx;
-
if (tx) {
crypto_info = &ctx->crypto_send;
cctx = &ctx->tx;
- aead = &sw_ctx->aead_send;
+ aead = &sw_ctx_tx->aead_send;
} else {
crypto_info = &ctx->crypto_recv;
cctx = &ctx->rx;
- aead = &sw_ctx->aead_recv;
+ aead = &sw_ctx_rx->aead_recv;
}
switch (crypto_info->cipher_type) {
@@ -1121,22 +1134,24 @@ int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx)
}
memcpy(cctx->rec_seq, rec_seq, rec_seq_size);
- if (tx) {
- sg_init_table(sw_ctx->sg_encrypted_data,
- ARRAY_SIZE(sw_ctx->sg_encrypted_data));
- sg_init_table(sw_ctx->sg_plaintext_data,
- ARRAY_SIZE(sw_ctx->sg_plaintext_data));
-
- sg_init_table(sw_ctx->sg_aead_in, 2);
- sg_set_buf(&sw_ctx->sg_aead_in[0], sw_ctx->aad_space,
- sizeof(sw_ctx->aad_space));
- sg_unmark_end(&sw_ctx->sg_aead_in[1]);
- sg_chain(sw_ctx->sg_aead_in, 2, sw_ctx->sg_plaintext_data);
- sg_init_table(sw_ctx->sg_aead_out, 2);
- sg_set_buf(&sw_ctx->sg_aead_out[0], sw_ctx->aad_space,
- sizeof(sw_ctx->aad_space));
- sg_unmark_end(&sw_ctx->sg_aead_out[1]);
- sg_chain(sw_ctx->sg_aead_out, 2, sw_ctx->sg_encrypted_data);
+ if (sw_ctx_tx) {
+ sg_init_table(sw_ctx_tx->sg_encrypted_data,
+ ARRAY_SIZE(sw_ctx_tx->sg_encrypted_data));
+ sg_init_table(sw_ctx_tx->sg_plaintext_data,
+ ARRAY_SIZE(sw_ctx_tx->sg_plaintext_data));
+
+ sg_init_table(sw_ctx_tx->sg_aead_in, 2);
+ sg_set_buf(&sw_ctx_tx->sg_aead_in[0], sw_ctx_tx->aad_space,
+ sizeof(sw_ctx_tx->aad_space));
+ sg_unmark_end(&sw_ctx_tx->sg_aead_in[1]);
+ sg_chain(sw_ctx_tx->sg_aead_in, 2,
+ sw_ctx_tx->sg_plaintext_data);
+ sg_init_table(sw_ctx_tx->sg_aead_out, 2);
+ sg_set_buf(&sw_ctx_tx->sg_aead_out[0], sw_ctx_tx->aad_space,
+ sizeof(sw_ctx_tx->aad_space));
+ sg_unmark_end(&sw_ctx_tx->sg_aead_out[1]);
+ sg_chain(sw_ctx_tx->sg_aead_out, 2,
+ sw_ctx_tx->sg_encrypted_data);
}
if (!*aead) {
@@ -1161,22 +1176,22 @@ int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx)
if (rc)
goto free_aead;
- if (!tx) {
+ if (sw_ctx_rx) {
/* Set up strparser */
memset(&cb, 0, sizeof(cb));
cb.rcv_msg = tls_queue;
cb.parse_msg = tls_read_size;
- strp_init(&sw_ctx->strp, sk, &cb);
+ strp_init(&sw_ctx_rx->strp, sk, &cb);
write_lock_bh(&sk->sk_callback_lock);
- sw_ctx->saved_data_ready = sk->sk_data_ready;
+ sw_ctx_rx->saved_data_ready = sk->sk_data_ready;
sk->sk_data_ready = tls_data_ready;
write_unlock_bh(&sk->sk_callback_lock);
- sw_ctx->sk_poll = sk->sk_socket->ops->poll;
+ sw_ctx_rx->sk_poll = sk->sk_socket->ops->poll;
- strp_check_rcv(&sw_ctx->strp);
+ strp_check_rcv(&sw_ctx_rx->strp);
}
goto out;
@@ -1188,11 +1203,16 @@ int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx)
kfree(cctx->rec_seq);
cctx->rec_seq = NULL;
free_iv:
- kfree(ctx->tx.iv);
- ctx->tx.iv = NULL;
+ kfree(cctx->iv);
+ cctx->iv = NULL;
free_priv:
- kfree(ctx->priv_ctx);
- ctx->priv_ctx = NULL;
+ if (tx) {
+ kfree(ctx->priv_ctx_tx);
+ ctx->priv_ctx_tx = NULL;
+ } else {
+ kfree(ctx->priv_ctx_rx);
+ ctx->priv_ctx_rx = NULL;
+ }
out:
return rc;
}
--
1.8.3.1
^ permalink raw reply related
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