mirror of
https://github.com/AuxXxilium/linux_dsm_epyc7002.git
synced 2024-12-04 03:46:42 +07:00
5a0e3ad6af
percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
355 lines
11 KiB
C
355 lines
11 KiB
C
/*
|
|
* WiMedia Logical Link Control Protocol (WLP)
|
|
* Message exchange infrastructure
|
|
*
|
|
* Copyright (C) 2007 Intel Corporation
|
|
* Reinette Chatre <reinette.chatre@intel.com>
|
|
*
|
|
* This program is free software; you can redistribute it and/or
|
|
* modify it under the terms of the GNU General Public License version
|
|
* 2 as published by the Free Software Foundation.
|
|
*
|
|
* This program is distributed in the hope that it will be useful,
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
* GNU General Public License for more details.
|
|
*
|
|
* You should have received a copy of the GNU General Public License
|
|
* along with this program; if not, write to the Free Software
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
|
|
* 02110-1301, USA.
|
|
*
|
|
*
|
|
* FIXME: Docs
|
|
*
|
|
*/
|
|
|
|
#include <linux/etherdevice.h>
|
|
#include <linux/slab.h>
|
|
#include <linux/wlp.h>
|
|
|
|
#include "wlp-internal.h"
|
|
|
|
/*
|
|
* Direct incoming association msg to correct parsing routine
|
|
*
|
|
* We only expect D1, E1, C1, C3 messages as new. All other incoming
|
|
* association messages should form part of an established session that is
|
|
* handled elsewhere.
|
|
* The handling of these messages often require calling sleeping functions
|
|
* - this cannot be done in interrupt context. We use the kernel's
|
|
* workqueue to handle these messages.
|
|
*/
|
|
static
|
|
void wlp_direct_assoc_frame(struct wlp *wlp, struct sk_buff *skb,
|
|
struct uwb_dev_addr *src)
|
|
{
|
|
struct device *dev = &wlp->rc->uwb_dev.dev;
|
|
struct wlp_frame_assoc *assoc = (void *) skb->data;
|
|
struct wlp_assoc_frame_ctx *frame_ctx;
|
|
|
|
frame_ctx = kmalloc(sizeof(*frame_ctx), GFP_ATOMIC);
|
|
if (frame_ctx == NULL) {
|
|
dev_err(dev, "WLP: Unable to allocate memory for association "
|
|
"frame handling.\n");
|
|
kfree_skb(skb);
|
|
return;
|
|
}
|
|
frame_ctx->wlp = wlp;
|
|
frame_ctx->skb = skb;
|
|
frame_ctx->src = *src;
|
|
switch (assoc->type) {
|
|
case WLP_ASSOC_D1:
|
|
INIT_WORK(&frame_ctx->ws, wlp_handle_d1_frame);
|
|
schedule_work(&frame_ctx->ws);
|
|
break;
|
|
case WLP_ASSOC_E1:
|
|
kfree_skb(skb); /* Temporary until we handle it */
|
|
kfree(frame_ctx); /* Temporary until we handle it */
|
|
break;
|
|
case WLP_ASSOC_C1:
|
|
INIT_WORK(&frame_ctx->ws, wlp_handle_c1_frame);
|
|
schedule_work(&frame_ctx->ws);
|
|
break;
|
|
case WLP_ASSOC_C3:
|
|
INIT_WORK(&frame_ctx->ws, wlp_handle_c3_frame);
|
|
schedule_work(&frame_ctx->ws);
|
|
break;
|
|
default:
|
|
dev_err(dev, "Received unexpected association frame. "
|
|
"Type = %d \n", assoc->type);
|
|
kfree_skb(skb);
|
|
kfree(frame_ctx);
|
|
break;
|
|
}
|
|
}
|
|
|
|
/*
|
|
* Process incoming association frame
|
|
*
|
|
* Although it could be possible to deal with some incoming association
|
|
* messages without creating a new session we are keeping things simple. We
|
|
* do not accept new association messages if there is a session in progress
|
|
* and the messages do not belong to that session.
|
|
*
|
|
* If an association message arrives that causes the creation of a session
|
|
* (WLP_ASSOC_E1) while we are in the process of creating a session then we
|
|
* rely on the neighbor mutex to protect the data. That is, the new session
|
|
* will not be started until the previous is completed.
|
|
*/
|
|
static
|
|
void wlp_receive_assoc_frame(struct wlp *wlp, struct sk_buff *skb,
|
|
struct uwb_dev_addr *src)
|
|
{
|
|
struct device *dev = &wlp->rc->uwb_dev.dev;
|
|
struct wlp_frame_assoc *assoc = (void *) skb->data;
|
|
struct wlp_session *session = wlp->session;
|
|
u8 version;
|
|
|
|
if (wlp_get_version(wlp, &assoc->version, &version,
|
|
sizeof(assoc->version)) < 0)
|
|
goto error;
|
|
if (version != WLP_VERSION) {
|
|
dev_err(dev, "Unsupported WLP version in association "
|
|
"message.\n");
|
|
goto error;
|
|
}
|
|
if (session != NULL) {
|
|
/* Function that created this session is still holding the
|
|
* &wlp->mutex to protect this session. */
|
|
if (assoc->type == session->exp_message ||
|
|
assoc->type == WLP_ASSOC_F0) {
|
|
if (!memcmp(&session->neighbor_addr, src,
|
|
sizeof(*src))) {
|
|
session->data = skb;
|
|
(session->cb)(wlp);
|
|
} else {
|
|
dev_err(dev, "Received expected message from "
|
|
"unexpected source. Expected message "
|
|
"%d or F0 from %02x:%02x, but received "
|
|
"it from %02x:%02x. Dropping.\n",
|
|
session->exp_message,
|
|
session->neighbor_addr.data[1],
|
|
session->neighbor_addr.data[0],
|
|
src->data[1], src->data[0]);
|
|
goto error;
|
|
}
|
|
} else {
|
|
dev_err(dev, "Association already in progress. "
|
|
"Dropping.\n");
|
|
goto error;
|
|
}
|
|
} else {
|
|
wlp_direct_assoc_frame(wlp, skb, src);
|
|
}
|
|
return;
|
|
error:
|
|
kfree_skb(skb);
|
|
}
|
|
|
|
/*
|
|
* Verify incoming frame is from connected neighbor, prep to pass to WLP client
|
|
*
|
|
* Verification proceeds according to WLP 0.99 [7.3.1]. The source address
|
|
* is used to determine which neighbor is sending the frame and the WSS tag
|
|
* is used to know to which WSS the frame belongs (we only support one WSS
|
|
* so this test is straight forward).
|
|
* With the WSS found we need to ensure that we are connected before
|
|
* allowing the exchange of data frames.
|
|
*/
|
|
static
|
|
int wlp_verify_prep_rx_frame(struct wlp *wlp, struct sk_buff *skb,
|
|
struct uwb_dev_addr *src)
|
|
{
|
|
struct device *dev = &wlp->rc->uwb_dev.dev;
|
|
int result = -EINVAL;
|
|
struct wlp_eda_node eda_entry;
|
|
struct wlp_frame_std_abbrv_hdr *hdr = (void *) skb->data;
|
|
|
|
/*verify*/
|
|
result = wlp_copy_eda_node(&wlp->eda, src, &eda_entry);
|
|
if (result < 0) {
|
|
if (printk_ratelimit())
|
|
dev_err(dev, "WLP: Incoming frame is from unknown "
|
|
"neighbor %02x:%02x.\n", src->data[1],
|
|
src->data[0]);
|
|
goto out;
|
|
}
|
|
if (hdr->tag != eda_entry.tag) {
|
|
if (printk_ratelimit())
|
|
dev_err(dev, "WLP: Tag of incoming frame from "
|
|
"%02x:%02x does not match expected tag. "
|
|
"Received 0x%02x, expected 0x%02x. \n",
|
|
src->data[1], src->data[0], hdr->tag,
|
|
eda_entry.tag);
|
|
result = -EINVAL;
|
|
goto out;
|
|
}
|
|
if (eda_entry.state != WLP_WSS_CONNECTED) {
|
|
if (printk_ratelimit())
|
|
dev_err(dev, "WLP: Incoming frame from "
|
|
"%02x:%02x does is not from connected WSS.\n",
|
|
src->data[1], src->data[0]);
|
|
result = -EINVAL;
|
|
goto out;
|
|
}
|
|
/*prep*/
|
|
skb_pull(skb, sizeof(*hdr));
|
|
out:
|
|
return result;
|
|
}
|
|
|
|
/*
|
|
* Receive a WLP frame from device
|
|
*
|
|
* @returns: 1 if calling function should free the skb
|
|
* 0 if it successfully handled skb and freed it
|
|
* 0 if error occured, will free skb in this case
|
|
*/
|
|
int wlp_receive_frame(struct device *dev, struct wlp *wlp, struct sk_buff *skb,
|
|
struct uwb_dev_addr *src)
|
|
{
|
|
unsigned len = skb->len;
|
|
void *ptr = skb->data;
|
|
struct wlp_frame_hdr *hdr;
|
|
int result = 0;
|
|
|
|
if (len < sizeof(*hdr)) {
|
|
dev_err(dev, "Not enough data to parse WLP header.\n");
|
|
result = -EINVAL;
|
|
goto out;
|
|
}
|
|
hdr = ptr;
|
|
if (le16_to_cpu(hdr->mux_hdr) != WLP_PROTOCOL_ID) {
|
|
dev_err(dev, "Not a WLP frame type.\n");
|
|
result = -EINVAL;
|
|
goto out;
|
|
}
|
|
switch (hdr->type) {
|
|
case WLP_FRAME_STANDARD:
|
|
if (len < sizeof(struct wlp_frame_std_abbrv_hdr)) {
|
|
dev_err(dev, "Not enough data to parse Standard "
|
|
"WLP header.\n");
|
|
goto out;
|
|
}
|
|
result = wlp_verify_prep_rx_frame(wlp, skb, src);
|
|
if (result < 0) {
|
|
if (printk_ratelimit())
|
|
dev_err(dev, "WLP: Verification of frame "
|
|
"from neighbor %02x:%02x failed.\n",
|
|
src->data[1], src->data[0]);
|
|
goto out;
|
|
}
|
|
result = 1;
|
|
break;
|
|
case WLP_FRAME_ABBREVIATED:
|
|
dev_err(dev, "Abbreviated frame received. FIXME?\n");
|
|
kfree_skb(skb);
|
|
break;
|
|
case WLP_FRAME_CONTROL:
|
|
dev_err(dev, "Control frame received. FIXME?\n");
|
|
kfree_skb(skb);
|
|
break;
|
|
case WLP_FRAME_ASSOCIATION:
|
|
if (len < sizeof(struct wlp_frame_assoc)) {
|
|
dev_err(dev, "Not enough data to parse Association "
|
|
"WLP header.\n");
|
|
goto out;
|
|
}
|
|
wlp_receive_assoc_frame(wlp, skb, src);
|
|
break;
|
|
default:
|
|
dev_err(dev, "Invalid frame received.\n");
|
|
result = -EINVAL;
|
|
break;
|
|
}
|
|
out:
|
|
if (result < 0) {
|
|
kfree_skb(skb);
|
|
result = 0;
|
|
}
|
|
return result;
|
|
}
|
|
EXPORT_SYMBOL_GPL(wlp_receive_frame);
|
|
|
|
|
|
/*
|
|
* Verify frame from network stack, prepare for further transmission
|
|
*
|
|
* @skb: the socket buffer that needs to be prepared for transmission (it
|
|
* is in need of a WLP header). If this is a broadcast frame we take
|
|
* over the entire transmission.
|
|
* If it is a unicast the WSS connection should already be established
|
|
* and transmission will be done by the calling function.
|
|
* @dst: On return this will contain the device address to which the
|
|
* frame is destined.
|
|
* @returns: 0 on success no tx : WLP header successfully applied to skb buffer,
|
|
* calling function can proceed with tx
|
|
* 1 on success with tx : WLP will take over transmission of this
|
|
* frame
|
|
* <0 on error
|
|
*
|
|
* The network stack (WLP client) is attempting to transmit a frame. We can
|
|
* only transmit data if a local WSS is at least active (connection will be
|
|
* done here if this is a broadcast frame and neighbor also has the WSS
|
|
* active).
|
|
*
|
|
* The frame can be either broadcast or unicast. Broadcast in a WSS is
|
|
* supported via multicast, but we don't support multicast yet (until
|
|
* devices start to support MAB IEs). If a broadcast frame needs to be
|
|
* transmitted it is treated as a unicast frame to each neighbor. In this
|
|
* case the WLP takes over transmission of the skb and returns 1
|
|
* to the caller to indicate so. Also, in this case, if a neighbor has the
|
|
* same WSS activated but is not connected then the WSS connection will be
|
|
* done at this time. The neighbor's virtual address will be learned at
|
|
* this time.
|
|
*
|
|
* The destination address in a unicast frame is the virtual address of the
|
|
* neighbor. This address only becomes known when a WSS connection is
|
|
* established. We thus rely on a broadcast frame to trigger the setup of
|
|
* WSS connections to all neighbors before we are able to send unicast
|
|
* frames to them. This seems reasonable as IP would usually use ARP first
|
|
* before any unicast frames are sent.
|
|
*
|
|
* If we are already connected to the neighbor (neighbor's virtual address
|
|
* is known) we just prepare the WLP header and the caller will continue to
|
|
* send the frame.
|
|
*
|
|
* A failure in this function usually indicates something that cannot be
|
|
* fixed automatically. So, if this function fails (@return < 0) the calling
|
|
* function should not retry to send the frame as it will very likely keep
|
|
* failing.
|
|
*
|
|
*/
|
|
int wlp_prepare_tx_frame(struct device *dev, struct wlp *wlp,
|
|
struct sk_buff *skb, struct uwb_dev_addr *dst)
|
|
{
|
|
int result = -EINVAL;
|
|
struct ethhdr *eth_hdr = (void *) skb->data;
|
|
|
|
if (is_multicast_ether_addr(eth_hdr->h_dest)) {
|
|
result = wlp_eda_for_each(&wlp->eda, wlp_wss_send_copy, skb);
|
|
if (result < 0) {
|
|
if (printk_ratelimit())
|
|
dev_err(dev, "Unable to handle broadcast "
|
|
"frame from WLP client.\n");
|
|
goto out;
|
|
}
|
|
dev_kfree_skb_irq(skb);
|
|
result = 1;
|
|
/* Frame will be transmitted by WLP. */
|
|
} else {
|
|
result = wlp_eda_for_virtual(&wlp->eda, eth_hdr->h_dest, dst,
|
|
wlp_wss_prep_hdr, skb);
|
|
if (unlikely(result < 0)) {
|
|
if (printk_ratelimit())
|
|
dev_err(dev, "Unable to prepare "
|
|
"skb for transmission. \n");
|
|
goto out;
|
|
}
|
|
}
|
|
out:
|
|
return result;
|
|
}
|
|
EXPORT_SYMBOL_GPL(wlp_prepare_tx_frame);
|