1// SPDX-License-Identifier: GPL-2.0
2/* Copyright(c) 1999 - 2018 Intel Corporation. */
3
4#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
5
6#include <linux/module.h>
7#include <linux/types.h>
8#include <linux/init.h>
9#include <linux/pci.h>
10#include <linux/vmalloc.h>
11#include <linux/pagemap.h>
12#include <linux/delay.h>
13#include <linux/netdevice.h>
14#include <linux/interrupt.h>
15#include <linux/tcp.h>
16#include <linux/ipv6.h>
17#include <linux/slab.h>
18#include <net/checksum.h>
19#include <net/ip6_checksum.h>
20#include <linux/ethtool.h>
21#include <linux/if_vlan.h>
22#include <linux/cpu.h>
23#include <linux/smp.h>
24#include <linux/pm_qos.h>
25#include <linux/pm_runtime.h>
26#include <linux/aer.h>
27#include <linux/prefetch.h>
28
29#include "e1000.h"
30
31char e1000e_driver_name[] = "e1000e";
32
33#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV|NETIF_MSG_PROBE|NETIF_MSG_LINK)
34static int debug = -1;
35module_param(debug, int, 0);
36MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
37
38static const struct e1000_info *e1000_info_tbl[] = {
39	[board_82571]		= &e1000_82571_info,
40	[board_82572]		= &e1000_82572_info,
41	[board_82573]		= &e1000_82573_info,
42	[board_82574]		= &e1000_82574_info,
43	[board_82583]		= &e1000_82583_info,
44	[board_80003es2lan]	= &e1000_es2_info,
45	[board_ich8lan]		= &e1000_ich8_info,
46	[board_ich9lan]		= &e1000_ich9_info,
47	[board_ich10lan]	= &e1000_ich10_info,
48	[board_pchlan]		= &e1000_pch_info,
49	[board_pch2lan]		= &e1000_pch2_info,
50	[board_pch_lpt]		= &e1000_pch_lpt_info,
51	[board_pch_spt]		= &e1000_pch_spt_info,
52	[board_pch_cnp]		= &e1000_pch_cnp_info,
53	[board_pch_tgp]		= &e1000_pch_tgp_info,
54};
55
56struct e1000_reg_info {
57	u32 ofs;
58	char *name;
59};
60
61static const struct e1000_reg_info e1000_reg_info_tbl[] = {
62	/* General Registers */
63	{E1000_CTRL, "CTRL"},
64	{E1000_STATUS, "STATUS"},
65	{E1000_CTRL_EXT, "CTRL_EXT"},
66
67	/* Interrupt Registers */
68	{E1000_ICR, "ICR"},
69
70	/* Rx Registers */
71	{E1000_RCTL, "RCTL"},
72	{E1000_RDLEN(0), "RDLEN"},
73	{E1000_RDH(0), "RDH"},
74	{E1000_RDT(0), "RDT"},
75	{E1000_RDTR, "RDTR"},
76	{E1000_RXDCTL(0), "RXDCTL"},
77	{E1000_ERT, "ERT"},
78	{E1000_RDBAL(0), "RDBAL"},
79	{E1000_RDBAH(0), "RDBAH"},
80	{E1000_RDFH, "RDFH"},
81	{E1000_RDFT, "RDFT"},
82	{E1000_RDFHS, "RDFHS"},
83	{E1000_RDFTS, "RDFTS"},
84	{E1000_RDFPC, "RDFPC"},
85
86	/* Tx Registers */
87	{E1000_TCTL, "TCTL"},
88	{E1000_TDBAL(0), "TDBAL"},
89	{E1000_TDBAH(0), "TDBAH"},
90	{E1000_TDLEN(0), "TDLEN"},
91	{E1000_TDH(0), "TDH"},
92	{E1000_TDT(0), "TDT"},
93	{E1000_TIDV, "TIDV"},
94	{E1000_TXDCTL(0), "TXDCTL"},
95	{E1000_TADV, "TADV"},
96	{E1000_TARC(0), "TARC"},
97	{E1000_TDFH, "TDFH"},
98	{E1000_TDFT, "TDFT"},
99	{E1000_TDFHS, "TDFHS"},
100	{E1000_TDFTS, "TDFTS"},
101	{E1000_TDFPC, "TDFPC"},
102
103	/* List Terminator */
104	{0, NULL}
105};
106
107/**
108 * __ew32_prepare - prepare to write to MAC CSR register on certain parts
109 * @hw: pointer to the HW structure
110 *
111 * When updating the MAC CSR registers, the Manageability Engine (ME) could
112 * be accessing the registers at the same time.  Normally, this is handled in
113 * h/w by an arbiter but on some parts there is a bug that acknowledges Host
114 * accesses later than it should which could result in the register to have
115 * an incorrect value.  Workaround this by checking the FWSM register which
116 * has bit 24 set while ME is accessing MAC CSR registers, wait if it is set
117 * and try again a number of times.
118 **/
119static void __ew32_prepare(struct e1000_hw *hw)
120{
121	s32 i = E1000_ICH_FWSM_PCIM2PCI_COUNT;
122
123	while ((er32(FWSM) & E1000_ICH_FWSM_PCIM2PCI) && --i)
124		udelay(50);
125}
126
127void __ew32(struct e1000_hw *hw, unsigned long reg, u32 val)
128{
129	if (hw->adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
130		__ew32_prepare(hw);
131
132	writel(val, hw->hw_addr + reg);
133}
134
135/**
136 * e1000_regdump - register printout routine
137 * @hw: pointer to the HW structure
138 * @reginfo: pointer to the register info table
139 **/
140static void e1000_regdump(struct e1000_hw *hw, struct e1000_reg_info *reginfo)
141{
142	int n = 0;
143	char rname[16];
144	u32 regs[8];
145
146	switch (reginfo->ofs) {
147	case E1000_RXDCTL(0):
148		for (n = 0; n < 2; n++)
149			regs[n] = __er32(hw, E1000_RXDCTL(n));
150		break;
151	case E1000_TXDCTL(0):
152		for (n = 0; n < 2; n++)
153			regs[n] = __er32(hw, E1000_TXDCTL(n));
154		break;
155	case E1000_TARC(0):
156		for (n = 0; n < 2; n++)
157			regs[n] = __er32(hw, E1000_TARC(n));
158		break;
159	default:
160		pr_info("%-15s %08x\n",
161			reginfo->name, __er32(hw, reginfo->ofs));
162		return;
163	}
164
165	snprintf(rname, 16, "%s%s", reginfo->name, "[0-1]");
166	pr_info("%-15s %08x %08x\n", rname, regs[0], regs[1]);
167}
168
169static void e1000e_dump_ps_pages(struct e1000_adapter *adapter,
170				 struct e1000_buffer *bi)
171{
172	int i;
173	struct e1000_ps_page *ps_page;
174
175	for (i = 0; i < adapter->rx_ps_pages; i++) {
176		ps_page = &bi->ps_pages[i];
177
178		if (ps_page->page) {
179			pr_info("packet dump for ps_page %d:\n", i);
180			print_hex_dump(KERN_INFO, "", DUMP_PREFIX_ADDRESS,
181				       16, 1, page_address(ps_page->page),
182				       PAGE_SIZE, true);
183		}
184	}
185}
186
187/**
188 * e1000e_dump - Print registers, Tx-ring and Rx-ring
189 * @adapter: board private structure
190 **/
191static void e1000e_dump(struct e1000_adapter *adapter)
192{
193	struct net_device *netdev = adapter->netdev;
194	struct e1000_hw *hw = &adapter->hw;
195	struct e1000_reg_info *reginfo;
196	struct e1000_ring *tx_ring = adapter->tx_ring;
197	struct e1000_tx_desc *tx_desc;
198	struct my_u0 {
199		__le64 a;
200		__le64 b;
201	} *u0;
202	struct e1000_buffer *buffer_info;
203	struct e1000_ring *rx_ring = adapter->rx_ring;
204	union e1000_rx_desc_packet_split *rx_desc_ps;
205	union e1000_rx_desc_extended *rx_desc;
206	struct my_u1 {
207		__le64 a;
208		__le64 b;
209		__le64 c;
210		__le64 d;
211	} *u1;
212	u32 staterr;
213	int i = 0;
214
215	if (!netif_msg_hw(adapter))
216		return;
217
218	/* Print netdevice Info */
219	if (netdev) {
220		dev_info(&adapter->pdev->dev, "Net device Info\n");
221		pr_info("Device Name     state            trans_start\n");
222		pr_info("%-15s %016lX %016lX\n", netdev->name,
223			netdev->state, dev_trans_start(netdev));
224	}
225
226	/* Print Registers */
227	dev_info(&adapter->pdev->dev, "Register Dump\n");
228	pr_info(" Register Name   Value\n");
229	for (reginfo = (struct e1000_reg_info *)e1000_reg_info_tbl;
230	     reginfo->name; reginfo++) {
231		e1000_regdump(hw, reginfo);
232	}
233
234	/* Print Tx Ring Summary */
235	if (!netdev || !netif_running(netdev))
236		return;
237
238	dev_info(&adapter->pdev->dev, "Tx Ring Summary\n");
239	pr_info("Queue [NTU] [NTC] [bi(ntc)->dma  ] leng ntw timestamp\n");
240	buffer_info = &tx_ring->buffer_info[tx_ring->next_to_clean];
241	pr_info(" %5d %5X %5X %016llX %04X %3X %016llX\n",
242		0, tx_ring->next_to_use, tx_ring->next_to_clean,
243		(unsigned long long)buffer_info->dma,
244		buffer_info->length,
245		buffer_info->next_to_watch,
246		(unsigned long long)buffer_info->time_stamp);
247
248	/* Print Tx Ring */
249	if (!netif_msg_tx_done(adapter))
250		goto rx_ring_summary;
251
252	dev_info(&adapter->pdev->dev, "Tx Ring Dump\n");
253
254	/* Transmit Descriptor Formats - DEXT[29] is 0 (Legacy) or 1 (Extended)
255	 *
256	 * Legacy Transmit Descriptor
257	 *   +--------------------------------------------------------------+
258	 * 0 |         Buffer Address [63:0] (Reserved on Write Back)       |
259	 *   +--------------------------------------------------------------+
260	 * 8 | Special  |    CSS     | Status |  CMD    |  CSO   |  Length  |
261	 *   +--------------------------------------------------------------+
262	 *   63       48 47        36 35    32 31     24 23    16 15        0
263	 *
264	 * Extended Context Descriptor (DTYP=0x0) for TSO or checksum offload
265	 *   63      48 47    40 39       32 31             16 15    8 7      0
266	 *   +----------------------------------------------------------------+
267	 * 0 |  TUCSE  | TUCS0  |   TUCSS   |     IPCSE       | IPCS0 | IPCSS |
268	 *   +----------------------------------------------------------------+
269	 * 8 |   MSS   | HDRLEN | RSV | STA | TUCMD | DTYP |      PAYLEN      |
270	 *   +----------------------------------------------------------------+
271	 *   63      48 47    40 39 36 35 32 31   24 23  20 19                0
272	 *
273	 * Extended Data Descriptor (DTYP=0x1)
274	 *   +----------------------------------------------------------------+
275	 * 0 |                     Buffer Address [63:0]                      |
276	 *   +----------------------------------------------------------------+
277	 * 8 | VLAN tag |  POPTS  | Rsvd | Status | Command | DTYP |  DTALEN  |
278	 *   +----------------------------------------------------------------+
279	 *   63       48 47     40 39  36 35    32 31     24 23  20 19        0
280	 */
281	pr_info("Tl[desc]     [address 63:0  ] [SpeCssSCmCsLen] [bi->dma       ] leng  ntw timestamp        bi->skb <-- Legacy format\n");
282	pr_info("Tc[desc]     [Ce CoCsIpceCoS] [MssHlRSCm0Plen] [bi->dma       ] leng  ntw timestamp        bi->skb <-- Ext Context format\n");
283	pr_info("Td[desc]     [address 63:0  ] [VlaPoRSCm1Dlen] [bi->dma       ] leng  ntw timestamp        bi->skb <-- Ext Data format\n");
284	for (i = 0; tx_ring->desc && (i < tx_ring->count); i++) {
285		const char *next_desc;
286		tx_desc = E1000_TX_DESC(*tx_ring, i);
287		buffer_info = &tx_ring->buffer_info[i];
288		u0 = (struct my_u0 *)tx_desc;
289		if (i == tx_ring->next_to_use && i == tx_ring->next_to_clean)
290			next_desc = " NTC/U";
291		else if (i == tx_ring->next_to_use)
292			next_desc = " NTU";
293		else if (i == tx_ring->next_to_clean)
294			next_desc = " NTC";
295		else
296			next_desc = "";
297		pr_info("T%c[0x%03X]    %016llX %016llX %016llX %04X  %3X %016llX %p%s\n",
298			(!(le64_to_cpu(u0->b) & BIT(29)) ? 'l' :
299			 ((le64_to_cpu(u0->b) & BIT(20)) ? 'd' : 'c')),
300			i,
301			(unsigned long long)le64_to_cpu(u0->a),
302			(unsigned long long)le64_to_cpu(u0->b),
303			(unsigned long long)buffer_info->dma,
304			buffer_info->length, buffer_info->next_to_watch,
305			(unsigned long long)buffer_info->time_stamp,
306			buffer_info->skb, next_desc);
307
308		if (netif_msg_pktdata(adapter) && buffer_info->skb)
309			print_hex_dump(KERN_INFO, "", DUMP_PREFIX_ADDRESS,
310				       16, 1, buffer_info->skb->data,
311				       buffer_info->skb->len, true);
312	}
313
314	/* Print Rx Ring Summary */
315rx_ring_summary:
316	dev_info(&adapter->pdev->dev, "Rx Ring Summary\n");
317	pr_info("Queue [NTU] [NTC]\n");
318	pr_info(" %5d %5X %5X\n",
319		0, rx_ring->next_to_use, rx_ring->next_to_clean);
320
321	/* Print Rx Ring */
322	if (!netif_msg_rx_status(adapter))
323		return;
324
325	dev_info(&adapter->pdev->dev, "Rx Ring Dump\n");
326	switch (adapter->rx_ps_pages) {
327	case 1:
328	case 2:
329	case 3:
330		/* [Extended] Packet Split Receive Descriptor Format
331		 *
332		 *    +-----------------------------------------------------+
333		 *  0 |                Buffer Address 0 [63:0]              |
334		 *    +-----------------------------------------------------+
335		 *  8 |                Buffer Address 1 [63:0]              |
336		 *    +-----------------------------------------------------+
337		 * 16 |                Buffer Address 2 [63:0]              |
338		 *    +-----------------------------------------------------+
339		 * 24 |                Buffer Address 3 [63:0]              |
340		 *    +-----------------------------------------------------+
341		 */
342		pr_info("R  [desc]      [buffer 0 63:0 ] [buffer 1 63:0 ] [buffer 2 63:0 ] [buffer 3 63:0 ] [bi->dma       ] [bi->skb] <-- Ext Pkt Split format\n");
343		/* [Extended] Receive Descriptor (Write-Back) Format
344		 *
345		 *   63       48 47    32 31     13 12    8 7    4 3        0
346		 *   +------------------------------------------------------+
347		 * 0 | Packet   | IP     |  Rsvd   | MRQ   | Rsvd | MRQ RSS |
348		 *   | Checksum | Ident  |         | Queue |      |  Type   |
349		 *   +------------------------------------------------------+
350		 * 8 | VLAN Tag | Length | Extended Error | Extended Status |
351		 *   +------------------------------------------------------+
352		 *   63       48 47    32 31            20 19               0
353		 */
354		pr_info("RWB[desc]      [ck ipid mrqhsh] [vl   l0 ee  es] [ l3  l2  l1 hs] [reserved      ] ---------------- [bi->skb] <-- Ext Rx Write-Back format\n");
355		for (i = 0; i < rx_ring->count; i++) {
356			const char *next_desc;
357			buffer_info = &rx_ring->buffer_info[i];
358			rx_desc_ps = E1000_RX_DESC_PS(*rx_ring, i);
359			u1 = (struct my_u1 *)rx_desc_ps;
360			staterr =
361			    le32_to_cpu(rx_desc_ps->wb.middle.status_error);
362
363			if (i == rx_ring->next_to_use)
364				next_desc = " NTU";
365			else if (i == rx_ring->next_to_clean)
366				next_desc = " NTC";
367			else
368				next_desc = "";
369
370			if (staterr & E1000_RXD_STAT_DD) {
371				/* Descriptor Done */
372				pr_info("%s[0x%03X]     %016llX %016llX %016llX %016llX ---------------- %p%s\n",
373					"RWB", i,
374					(unsigned long long)le64_to_cpu(u1->a),
375					(unsigned long long)le64_to_cpu(u1->b),
376					(unsigned long long)le64_to_cpu(u1->c),
377					(unsigned long long)le64_to_cpu(u1->d),
378					buffer_info->skb, next_desc);
379			} else {
380				pr_info("%s[0x%03X]     %016llX %016llX %016llX %016llX %016llX %p%s\n",
381					"R  ", i,
382					(unsigned long long)le64_to_cpu(u1->a),
383					(unsigned long long)le64_to_cpu(u1->b),
384					(unsigned long long)le64_to_cpu(u1->c),
385					(unsigned long long)le64_to_cpu(u1->d),
386					(unsigned long long)buffer_info->dma,
387					buffer_info->skb, next_desc);
388
389				if (netif_msg_pktdata(adapter))
390					e1000e_dump_ps_pages(adapter,
391							     buffer_info);
392			}
393		}
394		break;
395	default:
396	case 0:
397		/* Extended Receive Descriptor (Read) Format
398		 *
399		 *   +-----------------------------------------------------+
400		 * 0 |                Buffer Address [63:0]                |
401		 *   +-----------------------------------------------------+
402		 * 8 |                      Reserved                       |
403		 *   +-----------------------------------------------------+
404		 */
405		pr_info("R  [desc]      [buf addr 63:0 ] [reserved 63:0 ] [bi->dma       ] [bi->skb] <-- Ext (Read) format\n");
406		/* Extended Receive Descriptor (Write-Back) Format
407		 *
408		 *   63       48 47    32 31    24 23            4 3        0
409		 *   +------------------------------------------------------+
410		 *   |     RSS Hash      |        |               |         |
411		 * 0 +-------------------+  Rsvd  |   Reserved    | MRQ RSS |
412		 *   | Packet   | IP     |        |               |  Type   |
413		 *   | Checksum | Ident  |        |               |         |
414		 *   +------------------------------------------------------+
415		 * 8 | VLAN Tag | Length | Extended Error | Extended Status |
416		 *   +------------------------------------------------------+
417		 *   63       48 47    32 31            20 19               0
418		 */
419		pr_info("RWB[desc]      [cs ipid    mrq] [vt   ln xe  xs] [bi->skb] <-- Ext (Write-Back) format\n");
420
421		for (i = 0; i < rx_ring->count; i++) {
422			const char *next_desc;
423
424			buffer_info = &rx_ring->buffer_info[i];
425			rx_desc = E1000_RX_DESC_EXT(*rx_ring, i);
426			u1 = (struct my_u1 *)rx_desc;
427			staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
428
429			if (i == rx_ring->next_to_use)
430				next_desc = " NTU";
431			else if (i == rx_ring->next_to_clean)
432				next_desc = " NTC";
433			else
434				next_desc = "";
435
436			if (staterr & E1000_RXD_STAT_DD) {
437				/* Descriptor Done */
438				pr_info("%s[0x%03X]     %016llX %016llX ---------------- %p%s\n",
439					"RWB", i,
440					(unsigned long long)le64_to_cpu(u1->a),
441					(unsigned long long)le64_to_cpu(u1->b),
442					buffer_info->skb, next_desc);
443			} else {
444				pr_info("%s[0x%03X]     %016llX %016llX %016llX %p%s\n",
445					"R  ", i,
446					(unsigned long long)le64_to_cpu(u1->a),
447					(unsigned long long)le64_to_cpu(u1->b),
448					(unsigned long long)buffer_info->dma,
449					buffer_info->skb, next_desc);
450
451				if (netif_msg_pktdata(adapter) &&
452				    buffer_info->skb)
453					print_hex_dump(KERN_INFO, "",
454						       DUMP_PREFIX_ADDRESS, 16,
455						       1,
456						       buffer_info->skb->data,
457						       adapter->rx_buffer_len,
458						       true);
459			}
460		}
461	}
462}
463
464/**
465 * e1000_desc_unused - calculate if we have unused descriptors
466 * @ring: pointer to ring struct to perform calculation on
467 **/
468static int e1000_desc_unused(struct e1000_ring *ring)
469{
470	if (ring->next_to_clean > ring->next_to_use)
471		return ring->next_to_clean - ring->next_to_use - 1;
472
473	return ring->count + ring->next_to_clean - ring->next_to_use - 1;
474}
475
476/**
477 * e1000e_systim_to_hwtstamp - convert system time value to hw time stamp
478 * @adapter: board private structure
479 * @hwtstamps: time stamp structure to update
480 * @systim: unsigned 64bit system time value.
481 *
482 * Convert the system time value stored in the RX/TXSTMP registers into a
483 * hwtstamp which can be used by the upper level time stamping functions.
484 *
485 * The 'systim_lock' spinlock is used to protect the consistency of the
486 * system time value. This is needed because reading the 64 bit time
487 * value involves reading two 32 bit registers. The first read latches the
488 * value.
489 **/
490static void e1000e_systim_to_hwtstamp(struct e1000_adapter *adapter,
491				      struct skb_shared_hwtstamps *hwtstamps,
492				      u64 systim)
493{
494	u64 ns;
495	unsigned long flags;
496
497	spin_lock_irqsave(&adapter->systim_lock, flags);
498	ns = timecounter_cyc2time(&adapter->tc, systim);
499	spin_unlock_irqrestore(&adapter->systim_lock, flags);
500
501	memset(hwtstamps, 0, sizeof(*hwtstamps));
502	hwtstamps->hwtstamp = ns_to_ktime(ns);
503}
504
505/**
506 * e1000e_rx_hwtstamp - utility function which checks for Rx time stamp
507 * @adapter: board private structure
508 * @status: descriptor extended error and status field
509 * @skb: particular skb to include time stamp
510 *
511 * If the time stamp is valid, convert it into the timecounter ns value
512 * and store that result into the shhwtstamps structure which is passed
513 * up the network stack.
514 **/
515static void e1000e_rx_hwtstamp(struct e1000_adapter *adapter, u32 status,
516			       struct sk_buff *skb)
517{
518	struct e1000_hw *hw = &adapter->hw;
519	u64 rxstmp;
520
521	if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP) ||
522	    !(status & E1000_RXDEXT_STATERR_TST) ||
523	    !(er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_VALID))
524		return;
525
526	/* The Rx time stamp registers contain the time stamp.  No other
527	 * received packet will be time stamped until the Rx time stamp
528	 * registers are read.  Because only one packet can be time stamped
529	 * at a time, the register values must belong to this packet and
530	 * therefore none of the other additional attributes need to be
531	 * compared.
532	 */
533	rxstmp = (u64)er32(RXSTMPL);
534	rxstmp |= (u64)er32(RXSTMPH) << 32;
535	e1000e_systim_to_hwtstamp(adapter, skb_hwtstamps(skb), rxstmp);
536
537	adapter->flags2 &= ~FLAG2_CHECK_RX_HWTSTAMP;
538}
539
540/**
541 * e1000_receive_skb - helper function to handle Rx indications
542 * @adapter: board private structure
543 * @netdev: pointer to netdev struct
544 * @staterr: descriptor extended error and status field as written by hardware
545 * @vlan: descriptor vlan field as written by hardware (no le/be conversion)
546 * @skb: pointer to sk_buff to be indicated to stack
547 **/
548static void e1000_receive_skb(struct e1000_adapter *adapter,
549			      struct net_device *netdev, struct sk_buff *skb,
550			      u32 staterr, __le16 vlan)
551{
552	u16 tag = le16_to_cpu(vlan);
553
554	e1000e_rx_hwtstamp(adapter, staterr, skb);
555
556	skb->protocol = eth_type_trans(skb, netdev);
557
558	if (staterr & E1000_RXD_STAT_VP)
559		__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q), tag);
560
561	napi_gro_receive(&adapter->napi, skb);
562}
563
564/**
565 * e1000_rx_checksum - Receive Checksum Offload
566 * @adapter: board private structure
567 * @status_err: receive descriptor status and error fields
568 * @skb: socket buffer with received data
569 **/
570static void e1000_rx_checksum(struct e1000_adapter *adapter, u32 status_err,
571			      struct sk_buff *skb)
572{
573	u16 status = (u16)status_err;
574	u8 errors = (u8)(status_err >> 24);
575
576	skb_checksum_none_assert(skb);
577
578	/* Rx checksum disabled */
579	if (!(adapter->netdev->features & NETIF_F_RXCSUM))
580		return;
581
582	/* Ignore Checksum bit is set */
583	if (status & E1000_RXD_STAT_IXSM)
584		return;
585
586	/* TCP/UDP checksum error bit or IP checksum error bit is set */
587	if (errors & (E1000_RXD_ERR_TCPE | E1000_RXD_ERR_IPE)) {
588		/* let the stack verify checksum errors */
589		adapter->hw_csum_err++;
590		return;
591	}
592
593	/* TCP/UDP Checksum has not been calculated */
594	if (!(status & (E1000_RXD_STAT_TCPCS | E1000_RXD_STAT_UDPCS)))
595		return;
596
597	/* It must be a TCP or UDP packet with a valid checksum */
598	skb->ip_summed = CHECKSUM_UNNECESSARY;
599	adapter->hw_csum_good++;
600}
601
602static void e1000e_update_rdt_wa(struct e1000_ring *rx_ring, unsigned int i)
603{
604	struct e1000_adapter *adapter = rx_ring->adapter;
605	struct e1000_hw *hw = &adapter->hw;
606
607	__ew32_prepare(hw);
608	writel(i, rx_ring->tail);
609
610	if (unlikely(i != readl(rx_ring->tail))) {
611		u32 rctl = er32(RCTL);
612
613		ew32(RCTL, rctl & ~E1000_RCTL_EN);
614		e_err("ME firmware caused invalid RDT - resetting\n");
615		schedule_work(&adapter->reset_task);
616	}
617}
618
619static void e1000e_update_tdt_wa(struct e1000_ring *tx_ring, unsigned int i)
620{
621	struct e1000_adapter *adapter = tx_ring->adapter;
622	struct e1000_hw *hw = &adapter->hw;
623
624	__ew32_prepare(hw);
625	writel(i, tx_ring->tail);
626
627	if (unlikely(i != readl(tx_ring->tail))) {
628		u32 tctl = er32(TCTL);
629
630		ew32(TCTL, tctl & ~E1000_TCTL_EN);
631		e_err("ME firmware caused invalid TDT - resetting\n");
632		schedule_work(&adapter->reset_task);
633	}
634}
635
636/**
637 * e1000_alloc_rx_buffers - Replace used receive buffers
638 * @rx_ring: Rx descriptor ring
639 * @cleaned_count: number to reallocate
640 * @gfp: flags for allocation
641 **/
642static void e1000_alloc_rx_buffers(struct e1000_ring *rx_ring,
643				   int cleaned_count, gfp_t gfp)
644{
645	struct e1000_adapter *adapter = rx_ring->adapter;
646	struct net_device *netdev = adapter->netdev;
647	struct pci_dev *pdev = adapter->pdev;
648	union e1000_rx_desc_extended *rx_desc;
649	struct e1000_buffer *buffer_info;
650	struct sk_buff *skb;
651	unsigned int i;
652	unsigned int bufsz = adapter->rx_buffer_len;
653
654	i = rx_ring->next_to_use;
655	buffer_info = &rx_ring->buffer_info[i];
656
657	while (cleaned_count--) {
658		skb = buffer_info->skb;
659		if (skb) {
660			skb_trim(skb, 0);
661			goto map_skb;
662		}
663
664		skb = __netdev_alloc_skb_ip_align(netdev, bufsz, gfp);
665		if (!skb) {
666			/* Better luck next round */
667			adapter->alloc_rx_buff_failed++;
668			break;
669		}
670
671		buffer_info->skb = skb;
672map_skb:
673		buffer_info->dma = dma_map_single(&pdev->dev, skb->data,
674						  adapter->rx_buffer_len,
675						  DMA_FROM_DEVICE);
676		if (dma_mapping_error(&pdev->dev, buffer_info->dma)) {
677			dev_err(&pdev->dev, "Rx DMA map failed\n");
678			adapter->rx_dma_failed++;
679			break;
680		}
681
682		rx_desc = E1000_RX_DESC_EXT(*rx_ring, i);
683		rx_desc->read.buffer_addr = cpu_to_le64(buffer_info->dma);
684
685		if (unlikely(!(i & (E1000_RX_BUFFER_WRITE - 1)))) {
686			/* Force memory writes to complete before letting h/w
687			 * know there are new descriptors to fetch.  (Only
688			 * applicable for weak-ordered memory model archs,
689			 * such as IA-64).
690			 */
691			wmb();
692			if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
693				e1000e_update_rdt_wa(rx_ring, i);
694			else
695				writel(i, rx_ring->tail);
696		}
697		i++;
698		if (i == rx_ring->count)
699			i = 0;
700		buffer_info = &rx_ring->buffer_info[i];
701	}
702
703	rx_ring->next_to_use = i;
704}
705
706/**
707 * e1000_alloc_rx_buffers_ps - Replace used receive buffers; packet split
708 * @rx_ring: Rx descriptor ring
709 * @cleaned_count: number to reallocate
710 * @gfp: flags for allocation
711 **/
712static void e1000_alloc_rx_buffers_ps(struct e1000_ring *rx_ring,
713				      int cleaned_count, gfp_t gfp)
714{
715	struct e1000_adapter *adapter = rx_ring->adapter;
716	struct net_device *netdev = adapter->netdev;
717	struct pci_dev *pdev = adapter->pdev;
718	union e1000_rx_desc_packet_split *rx_desc;
719	struct e1000_buffer *buffer_info;
720	struct e1000_ps_page *ps_page;
721	struct sk_buff *skb;
722	unsigned int i, j;
723
724	i = rx_ring->next_to_use;
725	buffer_info = &rx_ring->buffer_info[i];
726
727	while (cleaned_count--) {
728		rx_desc = E1000_RX_DESC_PS(*rx_ring, i);
729
730		for (j = 0; j < PS_PAGE_BUFFERS; j++) {
731			ps_page = &buffer_info->ps_pages[j];
732			if (j >= adapter->rx_ps_pages) {
733				/* all unused desc entries get hw null ptr */
734				rx_desc->read.buffer_addr[j + 1] =
735				    ~cpu_to_le64(0);
736				continue;
737			}
738			if (!ps_page->page) {
739				ps_page->page = alloc_page(gfp);
740				if (!ps_page->page) {
741					adapter->alloc_rx_buff_failed++;
742					goto no_buffers;
743				}
744				ps_page->dma = dma_map_page(&pdev->dev,
745							    ps_page->page,
746							    0, PAGE_SIZE,
747							    DMA_FROM_DEVICE);
748				if (dma_mapping_error(&pdev->dev,
749						      ps_page->dma)) {
750					dev_err(&adapter->pdev->dev,
751						"Rx DMA page map failed\n");
752					adapter->rx_dma_failed++;
753					goto no_buffers;
754				}
755			}
756			/* Refresh the desc even if buffer_addrs
757			 * didn't change because each write-back
758			 * erases this info.
759			 */
760			rx_desc->read.buffer_addr[j + 1] =
761			    cpu_to_le64(ps_page->dma);
762		}
763
764		skb = __netdev_alloc_skb_ip_align(netdev, adapter->rx_ps_bsize0,
765						  gfp);
766
767		if (!skb) {
768			adapter->alloc_rx_buff_failed++;
769			break;
770		}
771
772		buffer_info->skb = skb;
773		buffer_info->dma = dma_map_single(&pdev->dev, skb->data,
774						  adapter->rx_ps_bsize0,
775						  DMA_FROM_DEVICE);
776		if (dma_mapping_error(&pdev->dev, buffer_info->dma)) {
777			dev_err(&pdev->dev, "Rx DMA map failed\n");
778			adapter->rx_dma_failed++;
779			/* cleanup skb */
780			dev_kfree_skb_any(skb);
781			buffer_info->skb = NULL;
782			break;
783		}
784
785		rx_desc->read.buffer_addr[0] = cpu_to_le64(buffer_info->dma);
786
787		if (unlikely(!(i & (E1000_RX_BUFFER_WRITE - 1)))) {
788			/* Force memory writes to complete before letting h/w
789			 * know there are new descriptors to fetch.  (Only
790			 * applicable for weak-ordered memory model archs,
791			 * such as IA-64).
792			 */
793			wmb();
794			if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
795				e1000e_update_rdt_wa(rx_ring, i << 1);
796			else
797				writel(i << 1, rx_ring->tail);
798		}
799
800		i++;
801		if (i == rx_ring->count)
802			i = 0;
803		buffer_info = &rx_ring->buffer_info[i];
804	}
805
806no_buffers:
807	rx_ring->next_to_use = i;
808}
809
810/**
811 * e1000_alloc_jumbo_rx_buffers - Replace used jumbo receive buffers
812 * @rx_ring: Rx descriptor ring
813 * @cleaned_count: number of buffers to allocate this pass
814 * @gfp: flags for allocation
815 **/
816
817static void e1000_alloc_jumbo_rx_buffers(struct e1000_ring *rx_ring,
818					 int cleaned_count, gfp_t gfp)
819{
820	struct e1000_adapter *adapter = rx_ring->adapter;
821	struct net_device *netdev = adapter->netdev;
822	struct pci_dev *pdev = adapter->pdev;
823	union e1000_rx_desc_extended *rx_desc;
824	struct e1000_buffer *buffer_info;
825	struct sk_buff *skb;
826	unsigned int i;
827	unsigned int bufsz = 256 - 16;	/* for skb_reserve */
828
829	i = rx_ring->next_to_use;
830	buffer_info = &rx_ring->buffer_info[i];
831
832	while (cleaned_count--) {
833		skb = buffer_info->skb;
834		if (skb) {
835			skb_trim(skb, 0);
836			goto check_page;
837		}
838
839		skb = __netdev_alloc_skb_ip_align(netdev, bufsz, gfp);
840		if (unlikely(!skb)) {
841			/* Better luck next round */
842			adapter->alloc_rx_buff_failed++;
843			break;
844		}
845
846		buffer_info->skb = skb;
847check_page:
848		/* allocate a new page if necessary */
849		if (!buffer_info->page) {
850			buffer_info->page = alloc_page(gfp);
851			if (unlikely(!buffer_info->page)) {
852				adapter->alloc_rx_buff_failed++;
853				break;
854			}
855		}
856
857		if (!buffer_info->dma) {
858			buffer_info->dma = dma_map_page(&pdev->dev,
859							buffer_info->page, 0,
860							PAGE_SIZE,
861							DMA_FROM_DEVICE);
862			if (dma_mapping_error(&pdev->dev, buffer_info->dma)) {
863				adapter->alloc_rx_buff_failed++;
864				break;
865			}
866		}
867
868		rx_desc = E1000_RX_DESC_EXT(*rx_ring, i);
869		rx_desc->read.buffer_addr = cpu_to_le64(buffer_info->dma);
870
871		if (unlikely(++i == rx_ring->count))
872			i = 0;
873		buffer_info = &rx_ring->buffer_info[i];
874	}
875
876	if (likely(rx_ring->next_to_use != i)) {
877		rx_ring->next_to_use = i;
878		if (unlikely(i-- == 0))
879			i = (rx_ring->count - 1);
880
881		/* Force memory writes to complete before letting h/w
882		 * know there are new descriptors to fetch.  (Only
883		 * applicable for weak-ordered memory model archs,
884		 * such as IA-64).
885		 */
886		wmb();
887		if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
888			e1000e_update_rdt_wa(rx_ring, i);
889		else
890			writel(i, rx_ring->tail);
891	}
892}
893
894static inline void e1000_rx_hash(struct net_device *netdev, __le32 rss,
895				 struct sk_buff *skb)
896{
897	if (netdev->features & NETIF_F_RXHASH)
898		skb_set_hash(skb, le32_to_cpu(rss), PKT_HASH_TYPE_L3);
899}
900
901/**
902 * e1000_clean_rx_irq - Send received data up the network stack
903 * @rx_ring: Rx descriptor ring
904 * @work_done: output parameter for indicating completed work
905 * @work_to_do: how many packets we can clean
906 *
907 * the return value indicates whether actual cleaning was done, there
908 * is no guarantee that everything was cleaned
909 **/
910static bool e1000_clean_rx_irq(struct e1000_ring *rx_ring, int *work_done,
911			       int work_to_do)
912{
913	struct e1000_adapter *adapter = rx_ring->adapter;
914	struct net_device *netdev = adapter->netdev;
915	struct pci_dev *pdev = adapter->pdev;
916	struct e1000_hw *hw = &adapter->hw;
917	union e1000_rx_desc_extended *rx_desc, *next_rxd;
918	struct e1000_buffer *buffer_info, *next_buffer;
919	u32 length, staterr;
920	unsigned int i;
921	int cleaned_count = 0;
922	bool cleaned = false;
923	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
924
925	i = rx_ring->next_to_clean;
926	rx_desc = E1000_RX_DESC_EXT(*rx_ring, i);
927	staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
928	buffer_info = &rx_ring->buffer_info[i];
929
930	while (staterr & E1000_RXD_STAT_DD) {
931		struct sk_buff *skb;
932
933		if (*work_done >= work_to_do)
934			break;
935		(*work_done)++;
936		dma_rmb();	/* read descriptor and rx_buffer_info after status DD */
937
938		skb = buffer_info->skb;
939		buffer_info->skb = NULL;
940
941		prefetch(skb->data - NET_IP_ALIGN);
942
943		i++;
944		if (i == rx_ring->count)
945			i = 0;
946		next_rxd = E1000_RX_DESC_EXT(*rx_ring, i);
947		prefetch(next_rxd);
948
949		next_buffer = &rx_ring->buffer_info[i];
950
951		cleaned = true;
952		cleaned_count++;
953		dma_unmap_single(&pdev->dev, buffer_info->dma,
954				 adapter->rx_buffer_len, DMA_FROM_DEVICE);
955		buffer_info->dma = 0;
956
957		length = le16_to_cpu(rx_desc->wb.upper.length);
958
959		/* !EOP means multiple descriptors were used to store a single
960		 * packet, if that's the case we need to toss it.  In fact, we
961		 * need to toss every packet with the EOP bit clear and the
962		 * next frame that _does_ have the EOP bit set, as it is by
963		 * definition only a frame fragment
964		 */
965		if (unlikely(!(staterr & E1000_RXD_STAT_EOP)))
966			adapter->flags2 |= FLAG2_IS_DISCARDING;
967
968		if (adapter->flags2 & FLAG2_IS_DISCARDING) {
969			/* All receives must fit into a single buffer */
970			e_dbg("Receive packet consumed multiple buffers\n");
971			/* recycle */
972			buffer_info->skb = skb;
973			if (staterr & E1000_RXD_STAT_EOP)
974				adapter->flags2 &= ~FLAG2_IS_DISCARDING;
975			goto next_desc;
976		}
977
978		if (unlikely((staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) &&
979			     !(netdev->features & NETIF_F_RXALL))) {
980			/* recycle */
981			buffer_info->skb = skb;
982			goto next_desc;
983		}
984
985		/* adjust length to remove Ethernet CRC */
986		if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) {
987			/* If configured to store CRC, don't subtract FCS,
988			 * but keep the FCS bytes out of the total_rx_bytes
989			 * counter
990			 */
991			if (netdev->features & NETIF_F_RXFCS)
992				total_rx_bytes -= 4;
993			else
994				length -= 4;
995		}
996
997		total_rx_bytes += length;
998		total_rx_packets++;
999
1000		/* code added for copybreak, this should improve
1001		 * performance for small packets with large amounts
1002		 * of reassembly being done in the stack
1003		 */
1004		if (length < copybreak) {
1005			struct sk_buff *new_skb =
1006				napi_alloc_skb(&adapter->napi, length);
1007			if (new_skb) {
1008				skb_copy_to_linear_data_offset(new_skb,
1009							       -NET_IP_ALIGN,
1010							       (skb->data -
1011								NET_IP_ALIGN),
1012							       (length +
1013								NET_IP_ALIGN));
1014				/* save the skb in buffer_info as good */
1015				buffer_info->skb = skb;
1016				skb = new_skb;
1017			}
1018			/* else just continue with the old one */
1019		}
1020		/* end copybreak code */
1021		skb_put(skb, length);
1022
1023		/* Receive Checksum Offload */
1024		e1000_rx_checksum(adapter, staterr, skb);
1025
1026		e1000_rx_hash(netdev, rx_desc->wb.lower.hi_dword.rss, skb);
1027
1028		e1000_receive_skb(adapter, netdev, skb, staterr,
1029				  rx_desc->wb.upper.vlan);
1030
1031next_desc:
1032		rx_desc->wb.upper.status_error &= cpu_to_le32(~0xFF);
1033
1034		/* return some buffers to hardware, one at a time is too slow */
1035		if (cleaned_count >= E1000_RX_BUFFER_WRITE) {
1036			adapter->alloc_rx_buf(rx_ring, cleaned_count,
1037					      GFP_ATOMIC);
1038			cleaned_count = 0;
1039		}
1040
1041		/* use prefetched values */
1042		rx_desc = next_rxd;
1043		buffer_info = next_buffer;
1044
1045		staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
1046	}
1047	rx_ring->next_to_clean = i;
1048
1049	cleaned_count = e1000_desc_unused(rx_ring);
1050	if (cleaned_count)
1051		adapter->alloc_rx_buf(rx_ring, cleaned_count, GFP_ATOMIC);
1052
1053	adapter->total_rx_bytes += total_rx_bytes;
1054	adapter->total_rx_packets += total_rx_packets;
1055	return cleaned;
1056}
1057
1058static void e1000_put_txbuf(struct e1000_ring *tx_ring,
1059			    struct e1000_buffer *buffer_info,
1060			    bool drop)
1061{
1062	struct e1000_adapter *adapter = tx_ring->adapter;
1063
1064	if (buffer_info->dma) {
1065		if (buffer_info->mapped_as_page)
1066			dma_unmap_page(&adapter->pdev->dev, buffer_info->dma,
1067				       buffer_info->length, DMA_TO_DEVICE);
1068		else
1069			dma_unmap_single(&adapter->pdev->dev, buffer_info->dma,
1070					 buffer_info->length, DMA_TO_DEVICE);
1071		buffer_info->dma = 0;
1072	}
1073	if (buffer_info->skb) {
1074		if (drop)
1075			dev_kfree_skb_any(buffer_info->skb);
1076		else
1077			dev_consume_skb_any(buffer_info->skb);
1078		buffer_info->skb = NULL;
1079	}
1080	buffer_info->time_stamp = 0;
1081}
1082
1083static void e1000_print_hw_hang(struct work_struct *work)
1084{
1085	struct e1000_adapter *adapter = container_of(work,
1086						     struct e1000_adapter,
1087						     print_hang_task);
1088	struct net_device *netdev = adapter->netdev;
1089	struct e1000_ring *tx_ring = adapter->tx_ring;
1090	unsigned int i = tx_ring->next_to_clean;
1091	unsigned int eop = tx_ring->buffer_info[i].next_to_watch;
1092	struct e1000_tx_desc *eop_desc = E1000_TX_DESC(*tx_ring, eop);
1093	struct e1000_hw *hw = &adapter->hw;
1094	u16 phy_status, phy_1000t_status, phy_ext_status;
1095	u16 pci_status;
1096
1097	if (test_bit(__E1000_DOWN, &adapter->state))
1098		return;
1099
1100	if (!adapter->tx_hang_recheck && (adapter->flags2 & FLAG2_DMA_BURST)) {
1101		/* May be block on write-back, flush and detect again
1102		 * flush pending descriptor writebacks to memory
1103		 */
1104		ew32(TIDV, adapter->tx_int_delay | E1000_TIDV_FPD);
1105		/* execute the writes immediately */
1106		e1e_flush();
1107		/* Due to rare timing issues, write to TIDV again to ensure
1108		 * the write is successful
1109		 */
1110		ew32(TIDV, adapter->tx_int_delay | E1000_TIDV_FPD);
1111		/* execute the writes immediately */
1112		e1e_flush();
1113		adapter->tx_hang_recheck = true;
1114		return;
1115	}
1116	adapter->tx_hang_recheck = false;
1117
1118	if (er32(TDH(0)) == er32(TDT(0))) {
1119		e_dbg("false hang detected, ignoring\n");
1120		return;
1121	}
1122
1123	/* Real hang detected */
1124	netif_stop_queue(netdev);
1125
1126	e1e_rphy(hw, MII_BMSR, &phy_status);
1127	e1e_rphy(hw, MII_STAT1000, &phy_1000t_status);
1128	e1e_rphy(hw, MII_ESTATUS, &phy_ext_status);
1129
1130	pci_read_config_word(adapter->pdev, PCI_STATUS, &pci_status);
1131
1132	/* detected Hardware unit hang */
1133	e_err("Detected Hardware Unit Hang:\n"
1134	      "  TDH                  <%x>\n"
1135	      "  TDT                  <%x>\n"
1136	      "  next_to_use          <%x>\n"
1137	      "  next_to_clean        <%x>\n"
1138	      "buffer_info[next_to_clean]:\n"
1139	      "  time_stamp           <%lx>\n"
1140	      "  next_to_watch        <%x>\n"
1141	      "  jiffies              <%lx>\n"
1142	      "  next_to_watch.status <%x>\n"
1143	      "MAC Status             <%x>\n"
1144	      "PHY Status             <%x>\n"
1145	      "PHY 1000BASE-T Status  <%x>\n"
1146	      "PHY Extended Status    <%x>\n"
1147	      "PCI Status             <%x>\n",
1148	      readl(tx_ring->head), readl(tx_ring->tail), tx_ring->next_to_use,
1149	      tx_ring->next_to_clean, tx_ring->buffer_info[eop].time_stamp,
1150	      eop, jiffies, eop_desc->upper.fields.status, er32(STATUS),
1151	      phy_status, phy_1000t_status, phy_ext_status, pci_status);
1152
1153	e1000e_dump(adapter);
1154
1155	/* Suggest workaround for known h/w issue */
1156	if ((hw->mac.type == e1000_pchlan) && (er32(CTRL) & E1000_CTRL_TFCE))
1157		e_err("Try turning off Tx pause (flow control) via ethtool\n");
1158}
1159
1160/**
1161 * e1000e_tx_hwtstamp_work - check for Tx time stamp
1162 * @work: pointer to work struct
1163 *
1164 * This work function polls the TSYNCTXCTL valid bit to determine when a
1165 * timestamp has been taken for the current stored skb.  The timestamp must
1166 * be for this skb because only one such packet is allowed in the queue.
1167 */
1168static void e1000e_tx_hwtstamp_work(struct work_struct *work)
1169{
1170	struct e1000_adapter *adapter = container_of(work, struct e1000_adapter,
1171						     tx_hwtstamp_work);
1172	struct e1000_hw *hw = &adapter->hw;
1173
1174	if (er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_VALID) {
1175		struct sk_buff *skb = adapter->tx_hwtstamp_skb;
1176		struct skb_shared_hwtstamps shhwtstamps;
1177		u64 txstmp;
1178
1179		txstmp = er32(TXSTMPL);
1180		txstmp |= (u64)er32(TXSTMPH) << 32;
1181
1182		e1000e_systim_to_hwtstamp(adapter, &shhwtstamps, txstmp);
1183
1184		/* Clear the global tx_hwtstamp_skb pointer and force writes
1185		 * prior to notifying the stack of a Tx timestamp.
1186		 */
1187		adapter->tx_hwtstamp_skb = NULL;
1188		wmb(); /* force write prior to skb_tstamp_tx */
1189
1190		skb_tstamp_tx(skb, &shhwtstamps);
1191		dev_consume_skb_any(skb);
1192	} else if (time_after(jiffies, adapter->tx_hwtstamp_start
1193			      + adapter->tx_timeout_factor * HZ)) {
1194		dev_kfree_skb_any(adapter->tx_hwtstamp_skb);
1195		adapter->tx_hwtstamp_skb = NULL;
1196		adapter->tx_hwtstamp_timeouts++;
1197		e_warn("clearing Tx timestamp hang\n");
1198	} else {
1199		/* reschedule to check later */
1200		schedule_work(&adapter->tx_hwtstamp_work);
1201	}
1202}
1203
1204/**
1205 * e1000_clean_tx_irq - Reclaim resources after transmit completes
1206 * @tx_ring: Tx descriptor ring
1207 *
1208 * the return value indicates whether actual cleaning was done, there
1209 * is no guarantee that everything was cleaned
1210 **/
1211static bool e1000_clean_tx_irq(struct e1000_ring *tx_ring)
1212{
1213	struct e1000_adapter *adapter = tx_ring->adapter;
1214	struct net_device *netdev = adapter->netdev;
1215	struct e1000_hw *hw = &adapter->hw;
1216	struct e1000_tx_desc *tx_desc, *eop_desc;
1217	struct e1000_buffer *buffer_info;
1218	unsigned int i, eop;
1219	unsigned int count = 0;
1220	unsigned int total_tx_bytes = 0, total_tx_packets = 0;
1221	unsigned int bytes_compl = 0, pkts_compl = 0;
1222
1223	i = tx_ring->next_to_clean;
1224	eop = tx_ring->buffer_info[i].next_to_watch;
1225	eop_desc = E1000_TX_DESC(*tx_ring, eop);
1226
1227	while ((eop_desc->upper.data & cpu_to_le32(E1000_TXD_STAT_DD)) &&
1228	       (count < tx_ring->count)) {
1229		bool cleaned = false;
1230
1231		dma_rmb();		/* read buffer_info after eop_desc */
1232		for (; !cleaned; count++) {
1233			tx_desc = E1000_TX_DESC(*tx_ring, i);
1234			buffer_info = &tx_ring->buffer_info[i];
1235			cleaned = (i == eop);
1236
1237			if (cleaned) {
1238				total_tx_packets += buffer_info->segs;
1239				total_tx_bytes += buffer_info->bytecount;
1240				if (buffer_info->skb) {
1241					bytes_compl += buffer_info->skb->len;
1242					pkts_compl++;
1243				}
1244			}
1245
1246			e1000_put_txbuf(tx_ring, buffer_info, false);
1247			tx_desc->upper.data = 0;
1248
1249			i++;
1250			if (i == tx_ring->count)
1251				i = 0;
1252		}
1253
1254		if (i == tx_ring->next_to_use)
1255			break;
1256		eop = tx_ring->buffer_info[i].next_to_watch;
1257		eop_desc = E1000_TX_DESC(*tx_ring, eop);
1258	}
1259
1260	tx_ring->next_to_clean = i;
1261
1262	netdev_completed_queue(netdev, pkts_compl, bytes_compl);
1263
1264#define TX_WAKE_THRESHOLD 32
1265	if (count && netif_carrier_ok(netdev) &&
1266	    e1000_desc_unused(tx_ring) >= TX_WAKE_THRESHOLD) {
1267		/* Make sure that anybody stopping the queue after this
1268		 * sees the new next_to_clean.
1269		 */
1270		smp_mb();
1271
1272		if (netif_queue_stopped(netdev) &&
1273		    !(test_bit(__E1000_DOWN, &adapter->state))) {
1274			netif_wake_queue(netdev);
1275			++adapter->restart_queue;
1276		}
1277	}
1278
1279	if (adapter->detect_tx_hung) {
1280		/* Detect a transmit hang in hardware, this serializes the
1281		 * check with the clearing of time_stamp and movement of i
1282		 */
1283		adapter->detect_tx_hung = false;
1284		if (tx_ring->buffer_info[i].time_stamp &&
1285		    time_after(jiffies, tx_ring->buffer_info[i].time_stamp
1286			       + (adapter->tx_timeout_factor * HZ)) &&
1287		    !(er32(STATUS) & E1000_STATUS_TXOFF))
1288			schedule_work(&adapter->print_hang_task);
1289		else
1290			adapter->tx_hang_recheck = false;
1291	}
1292	adapter->total_tx_bytes += total_tx_bytes;
1293	adapter->total_tx_packets += total_tx_packets;
1294	return count < tx_ring->count;
1295}
1296
1297/**
1298 * e1000_clean_rx_irq_ps - Send received data up the network stack; packet split
1299 * @rx_ring: Rx descriptor ring
1300 * @work_done: output parameter for indicating completed work
1301 * @work_to_do: how many packets we can clean
1302 *
1303 * the return value indicates whether actual cleaning was done, there
1304 * is no guarantee that everything was cleaned
1305 **/
1306static bool e1000_clean_rx_irq_ps(struct e1000_ring *rx_ring, int *work_done,
1307				  int work_to_do)
1308{
1309	struct e1000_adapter *adapter = rx_ring->adapter;
1310	struct e1000_hw *hw = &adapter->hw;
1311	union e1000_rx_desc_packet_split *rx_desc, *next_rxd;
1312	struct net_device *netdev = adapter->netdev;
1313	struct pci_dev *pdev = adapter->pdev;
1314	struct e1000_buffer *buffer_info, *next_buffer;
1315	struct e1000_ps_page *ps_page;
1316	struct sk_buff *skb;
1317	unsigned int i, j;
1318	u32 length, staterr;
1319	int cleaned_count = 0;
1320	bool cleaned = false;
1321	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
1322
1323	i = rx_ring->next_to_clean;
1324	rx_desc = E1000_RX_DESC_PS(*rx_ring, i);
1325	staterr = le32_to_cpu(rx_desc->wb.middle.status_error);
1326	buffer_info = &rx_ring->buffer_info[i];
1327
1328	while (staterr & E1000_RXD_STAT_DD) {
1329		if (*work_done >= work_to_do)
1330			break;
1331		(*work_done)++;
1332		skb = buffer_info->skb;
1333		dma_rmb();	/* read descriptor and rx_buffer_info after status DD */
1334
1335		/* in the packet split case this is header only */
1336		prefetch(skb->data - NET_IP_ALIGN);
1337
1338		i++;
1339		if (i == rx_ring->count)
1340			i = 0;
1341		next_rxd = E1000_RX_DESC_PS(*rx_ring, i);
1342		prefetch(next_rxd);
1343
1344		next_buffer = &rx_ring->buffer_info[i];
1345
1346		cleaned = true;
1347		cleaned_count++;
1348		dma_unmap_single(&pdev->dev, buffer_info->dma,
1349				 adapter->rx_ps_bsize0, DMA_FROM_DEVICE);
1350		buffer_info->dma = 0;
1351
1352		/* see !EOP comment in other Rx routine */
1353		if (!(staterr & E1000_RXD_STAT_EOP))
1354			adapter->flags2 |= FLAG2_IS_DISCARDING;
1355
1356		if (adapter->flags2 & FLAG2_IS_DISCARDING) {
1357			e_dbg("Packet Split buffers didn't pick up the full packet\n");
1358			dev_kfree_skb_irq(skb);
1359			if (staterr & E1000_RXD_STAT_EOP)
1360				adapter->flags2 &= ~FLAG2_IS_DISCARDING;
1361			goto next_desc;
1362		}
1363
1364		if (unlikely((staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) &&
1365			     !(netdev->features & NETIF_F_RXALL))) {
1366			dev_kfree_skb_irq(skb);
1367			goto next_desc;
1368		}
1369
1370		length = le16_to_cpu(rx_desc->wb.middle.length0);
1371
1372		if (!length) {
1373			e_dbg("Last part of the packet spanning multiple descriptors\n");
1374			dev_kfree_skb_irq(skb);
1375			goto next_desc;
1376		}
1377
1378		/* Good Receive */
1379		skb_put(skb, length);
1380
1381		{
1382			/* this looks ugly, but it seems compiler issues make
1383			 * it more efficient than reusing j
1384			 */
1385			int l1 = le16_to_cpu(rx_desc->wb.upper.length[0]);
1386
1387			/* page alloc/put takes too long and effects small
1388			 * packet throughput, so unsplit small packets and
1389			 * save the alloc/put only valid in softirq (napi)
1390			 * context to call kmap_*
1391			 */
1392			if (l1 && (l1 <= copybreak) &&
1393			    ((length + l1) <= adapter->rx_ps_bsize0)) {
1394				u8 *vaddr;
1395
1396				ps_page = &buffer_info->ps_pages[0];
1397
1398				/* there is no documentation about how to call
1399				 * kmap_atomic, so we can't hold the mapping
1400				 * very long
1401				 */
1402				dma_sync_single_for_cpu(&pdev->dev,
1403							ps_page->dma,
1404							PAGE_SIZE,
1405							DMA_FROM_DEVICE);
1406				vaddr = kmap_atomic(ps_page->page);
1407				memcpy(skb_tail_pointer(skb), vaddr, l1);
1408				kunmap_atomic(vaddr);
1409				dma_sync_single_for_device(&pdev->dev,
1410							   ps_page->dma,
1411							   PAGE_SIZE,
1412							   DMA_FROM_DEVICE);
1413
1414				/* remove the CRC */
1415				if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) {
1416					if (!(netdev->features & NETIF_F_RXFCS))
1417						l1 -= 4;
1418				}
1419
1420				skb_put(skb, l1);
1421				goto copydone;
1422			}	/* if */
1423		}
1424
1425		for (j = 0; j < PS_PAGE_BUFFERS; j++) {
1426			length = le16_to_cpu(rx_desc->wb.upper.length[j]);
1427			if (!length)
1428				break;
1429
1430			ps_page = &buffer_info->ps_pages[j];
1431			dma_unmap_page(&pdev->dev, ps_page->dma, PAGE_SIZE,
1432				       DMA_FROM_DEVICE);
1433			ps_page->dma = 0;
1434			skb_fill_page_desc(skb, j, ps_page->page, 0, length);
1435			ps_page->page = NULL;
1436			skb->len += length;
1437			skb->data_len += length;
1438			skb->truesize += PAGE_SIZE;
1439		}
1440
1441		/* strip the ethernet crc, problem is we're using pages now so
1442		 * this whole operation can get a little cpu intensive
1443		 */
1444		if (!(adapter->flags2 & FLAG2_CRC_STRIPPING)) {
1445			if (!(netdev->features & NETIF_F_RXFCS))
1446				pskb_trim(skb, skb->len - 4);
1447		}
1448
1449copydone:
1450		total_rx_bytes += skb->len;
1451		total_rx_packets++;
1452
1453		e1000_rx_checksum(adapter, staterr, skb);
1454
1455		e1000_rx_hash(netdev, rx_desc->wb.lower.hi_dword.rss, skb);
1456
1457		if (rx_desc->wb.upper.header_status &
1458		    cpu_to_le16(E1000_RXDPS_HDRSTAT_HDRSP))
1459			adapter->rx_hdr_split++;
1460
1461		e1000_receive_skb(adapter, netdev, skb, staterr,
1462				  rx_desc->wb.middle.vlan);
1463
1464next_desc:
1465		rx_desc->wb.middle.status_error &= cpu_to_le32(~0xFF);
1466		buffer_info->skb = NULL;
1467
1468		/* return some buffers to hardware, one at a time is too slow */
1469		if (cleaned_count >= E1000_RX_BUFFER_WRITE) {
1470			adapter->alloc_rx_buf(rx_ring, cleaned_count,
1471					      GFP_ATOMIC);
1472			cleaned_count = 0;
1473		}
1474
1475		/* use prefetched values */
1476		rx_desc = next_rxd;
1477		buffer_info = next_buffer;
1478
1479		staterr = le32_to_cpu(rx_desc->wb.middle.status_error);
1480	}
1481	rx_ring->next_to_clean = i;
1482
1483	cleaned_count = e1000_desc_unused(rx_ring);
1484	if (cleaned_count)
1485		adapter->alloc_rx_buf(rx_ring, cleaned_count, GFP_ATOMIC);
1486
1487	adapter->total_rx_bytes += total_rx_bytes;
1488	adapter->total_rx_packets += total_rx_packets;
1489	return cleaned;
1490}
1491
1492static void e1000_consume_page(struct e1000_buffer *bi, struct sk_buff *skb,
1493			       u16 length)
1494{
1495	bi->page = NULL;
1496	skb->len += length;
1497	skb->data_len += length;
1498	skb->truesize += PAGE_SIZE;
1499}
1500
1501/**
1502 * e1000_clean_jumbo_rx_irq - Send received data up the network stack; legacy
1503 * @rx_ring: Rx descriptor ring
1504 * @work_done: output parameter for indicating completed work
1505 * @work_to_do: how many packets we can clean
1506 *
1507 * the return value indicates whether actual cleaning was done, there
1508 * is no guarantee that everything was cleaned
1509 **/
1510static bool e1000_clean_jumbo_rx_irq(struct e1000_ring *rx_ring, int *work_done,
1511				     int work_to_do)
1512{
1513	struct e1000_adapter *adapter = rx_ring->adapter;
1514	struct net_device *netdev = adapter->netdev;
1515	struct pci_dev *pdev = adapter->pdev;
1516	union e1000_rx_desc_extended *rx_desc, *next_rxd;
1517	struct e1000_buffer *buffer_info, *next_buffer;
1518	u32 length, staterr;
1519	unsigned int i;
1520	int cleaned_count = 0;
1521	bool cleaned = false;
1522	unsigned int total_rx_bytes = 0, total_rx_packets = 0;
1523	struct skb_shared_info *shinfo;
1524
1525	i = rx_ring->next_to_clean;
1526	rx_desc = E1000_RX_DESC_EXT(*rx_ring, i);
1527	staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
1528	buffer_info = &rx_ring->buffer_info[i];
1529
1530	while (staterr & E1000_RXD_STAT_DD) {
1531		struct sk_buff *skb;
1532
1533		if (*work_done >= work_to_do)
1534			break;
1535		(*work_done)++;
1536		dma_rmb();	/* read descriptor and rx_buffer_info after status DD */
1537
1538		skb = buffer_info->skb;
1539		buffer_info->skb = NULL;
1540
1541		++i;
1542		if (i == rx_ring->count)
1543			i = 0;
1544		next_rxd = E1000_RX_DESC_EXT(*rx_ring, i);
1545		prefetch(next_rxd);
1546
1547		next_buffer = &rx_ring->buffer_info[i];
1548
1549		cleaned = true;
1550		cleaned_count++;
1551		dma_unmap_page(&pdev->dev, buffer_info->dma, PAGE_SIZE,
1552			       DMA_FROM_DEVICE);
1553		buffer_info->dma = 0;
1554
1555		length = le16_to_cpu(rx_desc->wb.upper.length);
1556
1557		/* errors is only valid for DD + EOP descriptors */
1558		if (unlikely((staterr & E1000_RXD_STAT_EOP) &&
1559			     ((staterr & E1000_RXDEXT_ERR_FRAME_ERR_MASK) &&
1560			      !(netdev->features & NETIF_F_RXALL)))) {
1561			/* recycle both page and skb */
1562			buffer_info->skb = skb;
1563			/* an error means any chain goes out the window too */
1564			if (rx_ring->rx_skb_top)
1565				dev_kfree_skb_irq(rx_ring->rx_skb_top);
1566			rx_ring->rx_skb_top = NULL;
1567			goto next_desc;
1568		}
1569#define rxtop (rx_ring->rx_skb_top)
1570		if (!(staterr & E1000_RXD_STAT_EOP)) {
1571			/* this descriptor is only the beginning (or middle) */
1572			if (!rxtop) {
1573				/* this is the beginning of a chain */
1574				rxtop = skb;
1575				skb_fill_page_desc(rxtop, 0, buffer_info->page,
1576						   0, length);
1577			} else {
1578				/* this is the middle of a chain */
1579				shinfo = skb_shinfo(rxtop);
1580				skb_fill_page_desc(rxtop, shinfo->nr_frags,
1581						   buffer_info->page, 0,
1582						   length);
1583				/* re-use the skb, only consumed the page */
1584				buffer_info->skb = skb;
1585			}
1586			e1000_consume_page(buffer_info, rxtop, length);
1587			goto next_desc;
1588		} else {
1589			if (rxtop) {
1590				/* end of the chain */
1591				shinfo = skb_shinfo(rxtop);
1592				skb_fill_page_desc(rxtop, shinfo->nr_frags,
1593						   buffer_info->page, 0,
1594						   length);
1595				/* re-use the current skb, we only consumed the
1596				 * page
1597				 */
1598				buffer_info->skb = skb;
1599				skb = rxtop;
1600				rxtop = NULL;
1601				e1000_consume_page(buffer_info, skb, length);
1602			} else {
1603				/* no chain, got EOP, this buf is the packet
1604				 * copybreak to save the put_page/alloc_page
1605				 */
1606				if (length <= copybreak &&
1607				    skb_tailroom(skb) >= length) {
1608					u8 *vaddr;
1609					vaddr = kmap_atomic(buffer_info->page);
1610					memcpy(skb_tail_pointer(skb), vaddr,
1611					       length);
1612					kunmap_atomic(vaddr);
1613					/* re-use the page, so don't erase
1614					 * buffer_info->page
1615					 */
1616					skb_put(skb, length);
1617				} else {
1618					skb_fill_page_desc(skb, 0,
1619							   buffer_info->page, 0,
1620							   length);
1621					e1000_consume_page(buffer_info, skb,
1622							   length);
1623				}
1624			}
1625		}
1626
1627		/* Receive Checksum Offload */
1628		e1000_rx_checksum(adapter, staterr, skb);
1629
1630		e1000_rx_hash(netdev, rx_desc->wb.lower.hi_dword.rss, skb);
1631
1632		/* probably a little skewed due to removing CRC */
1633		total_rx_bytes += skb->len;
1634		total_rx_packets++;
1635
1636		/* eth type trans needs skb->data to point to something */
1637		if (!pskb_may_pull(skb, ETH_HLEN)) {
1638			e_err("pskb_may_pull failed.\n");
1639			dev_kfree_skb_irq(skb);
1640			goto next_desc;
1641		}
1642
1643		e1000_receive_skb(adapter, netdev, skb, staterr,
1644				  rx_desc->wb.upper.vlan);
1645
1646next_desc:
1647		rx_desc->wb.upper.status_error &= cpu_to_le32(~0xFF);
1648
1649		/* return some buffers to hardware, one at a time is too slow */
1650		if (unlikely(cleaned_count >= E1000_RX_BUFFER_WRITE)) {
1651			adapter->alloc_rx_buf(rx_ring, cleaned_count,
1652					      GFP_ATOMIC);
1653			cleaned_count = 0;
1654		}
1655
1656		/* use prefetched values */
1657		rx_desc = next_rxd;
1658		buffer_info = next_buffer;
1659
1660		staterr = le32_to_cpu(rx_desc->wb.upper.status_error);
1661	}
1662	rx_ring->next_to_clean = i;
1663
1664	cleaned_count = e1000_desc_unused(rx_ring);
1665	if (cleaned_count)
1666		adapter->alloc_rx_buf(rx_ring, cleaned_count, GFP_ATOMIC);
1667
1668	adapter->total_rx_bytes += total_rx_bytes;
1669	adapter->total_rx_packets += total_rx_packets;
1670	return cleaned;
1671}
1672
1673/**
1674 * e1000_clean_rx_ring - Free Rx Buffers per Queue
1675 * @rx_ring: Rx descriptor ring
1676 **/
1677static void e1000_clean_rx_ring(struct e1000_ring *rx_ring)
1678{
1679	struct e1000_adapter *adapter = rx_ring->adapter;
1680	struct e1000_buffer *buffer_info;
1681	struct e1000_ps_page *ps_page;
1682	struct pci_dev *pdev = adapter->pdev;
1683	unsigned int i, j;
1684
1685	/* Free all the Rx ring sk_buffs */
1686	for (i = 0; i < rx_ring->count; i++) {
1687		buffer_info = &rx_ring->buffer_info[i];
1688		if (buffer_info->dma) {
1689			if (adapter->clean_rx == e1000_clean_rx_irq)
1690				dma_unmap_single(&pdev->dev, buffer_info->dma,
1691						 adapter->rx_buffer_len,
1692						 DMA_FROM_DEVICE);
1693			else if (adapter->clean_rx == e1000_clean_jumbo_rx_irq)
1694				dma_unmap_page(&pdev->dev, buffer_info->dma,
1695					       PAGE_SIZE, DMA_FROM_DEVICE);
1696			else if (adapter->clean_rx == e1000_clean_rx_irq_ps)
1697				dma_unmap_single(&pdev->dev, buffer_info->dma,
1698						 adapter->rx_ps_bsize0,
1699						 DMA_FROM_DEVICE);
1700			buffer_info->dma = 0;
1701		}
1702
1703		if (buffer_info->page) {
1704			put_page(buffer_info->page);
1705			buffer_info->page = NULL;
1706		}
1707
1708		if (buffer_info->skb) {
1709			dev_kfree_skb(buffer_info->skb);
1710			buffer_info->skb = NULL;
1711		}
1712
1713		for (j = 0; j < PS_PAGE_BUFFERS; j++) {
1714			ps_page = &buffer_info->ps_pages[j];
1715			if (!ps_page->page)
1716				break;
1717			dma_unmap_page(&pdev->dev, ps_page->dma, PAGE_SIZE,
1718				       DMA_FROM_DEVICE);
1719			ps_page->dma = 0;
1720			put_page(ps_page->page);
1721			ps_page->page = NULL;
1722		}
1723	}
1724
1725	/* there also may be some cached data from a chained receive */
1726	if (rx_ring->rx_skb_top) {
1727		dev_kfree_skb(rx_ring->rx_skb_top);
1728		rx_ring->rx_skb_top = NULL;
1729	}
1730
1731	/* Zero out the descriptor ring */
1732	memset(rx_ring->desc, 0, rx_ring->size);
1733
1734	rx_ring->next_to_clean = 0;
1735	rx_ring->next_to_use = 0;
1736	adapter->flags2 &= ~FLAG2_IS_DISCARDING;
1737}
1738
1739static void e1000e_downshift_workaround(struct work_struct *work)
1740{
1741	struct e1000_adapter *adapter = container_of(work,
1742						     struct e1000_adapter,
1743						     downshift_task);
1744
1745	if (test_bit(__E1000_DOWN, &adapter->state))
1746		return;
1747
1748	e1000e_gig_downshift_workaround_ich8lan(&adapter->hw);
1749}
1750
1751/**
1752 * e1000_intr_msi - Interrupt Handler
1753 * @irq: interrupt number
1754 * @data: pointer to a network interface device structure
1755 **/
1756static irqreturn_t e1000_intr_msi(int __always_unused irq, void *data)
1757{
1758	struct net_device *netdev = data;
1759	struct e1000_adapter *adapter = netdev_priv(netdev);
1760	struct e1000_hw *hw = &adapter->hw;
1761	u32 icr = er32(ICR);
1762
1763	/* read ICR disables interrupts using IAM */
1764	if (icr & E1000_ICR_LSC) {
1765		hw->mac.get_link_status = true;
1766		/* ICH8 workaround-- Call gig speed drop workaround on cable
1767		 * disconnect (LSC) before accessing any PHY registers
1768		 */
1769		if ((adapter->flags & FLAG_LSC_GIG_SPEED_DROP) &&
1770		    (!(er32(STATUS) & E1000_STATUS_LU)))
1771			schedule_work(&adapter->downshift_task);
1772
1773		/* 80003ES2LAN workaround-- For packet buffer work-around on
1774		 * link down event; disable receives here in the ISR and reset
1775		 * adapter in watchdog
1776		 */
1777		if (netif_carrier_ok(netdev) &&
1778		    adapter->flags & FLAG_RX_NEEDS_RESTART) {
1779			/* disable receives */
1780			u32 rctl = er32(RCTL);
1781
1782			ew32(RCTL, rctl & ~E1000_RCTL_EN);
1783			adapter->flags |= FLAG_RESTART_NOW;
1784		}
1785		/* guard against interrupt when we're going down */
1786		if (!test_bit(__E1000_DOWN, &adapter->state))
1787			mod_timer(&adapter->watchdog_timer, jiffies + 1);
1788	}
1789
1790	/* Reset on uncorrectable ECC error */
1791	if ((icr & E1000_ICR_ECCER) && (hw->mac.type >= e1000_pch_lpt)) {
1792		u32 pbeccsts = er32(PBECCSTS);
1793
1794		adapter->corr_errors +=
1795		    pbeccsts & E1000_PBECCSTS_CORR_ERR_CNT_MASK;
1796		adapter->uncorr_errors +=
1797		    (pbeccsts & E1000_PBECCSTS_UNCORR_ERR_CNT_MASK) >>
1798		    E1000_PBECCSTS_UNCORR_ERR_CNT_SHIFT;
1799
1800		/* Do the reset outside of interrupt context */
1801		schedule_work(&adapter->reset_task);
1802
1803		/* return immediately since reset is imminent */
1804		return IRQ_HANDLED;
1805	}
1806
1807	if (napi_schedule_prep(&adapter->napi)) {
1808		adapter->total_tx_bytes = 0;
1809		adapter->total_tx_packets = 0;
1810		adapter->total_rx_bytes = 0;
1811		adapter->total_rx_packets = 0;
1812		__napi_schedule(&adapter->napi);
1813	}
1814
1815	return IRQ_HANDLED;
1816}
1817
1818/**
1819 * e1000_intr - Interrupt Handler
1820 * @irq: interrupt number
1821 * @data: pointer to a network interface device structure
1822 **/
1823static irqreturn_t e1000_intr(int __always_unused irq, void *data)
1824{
1825	struct net_device *netdev = data;
1826	struct e1000_adapter *adapter = netdev_priv(netdev);
1827	struct e1000_hw *hw = &adapter->hw;
1828	u32 rctl, icr = er32(ICR);
1829
1830	if (!icr || test_bit(__E1000_DOWN, &adapter->state))
1831		return IRQ_NONE;	/* Not our interrupt */
1832
1833	/* IMS will not auto-mask if INT_ASSERTED is not set, and if it is
1834	 * not set, then the adapter didn't send an interrupt
1835	 */
1836	if (!(icr & E1000_ICR_INT_ASSERTED))
1837		return IRQ_NONE;
1838
1839	/* Interrupt Auto-Mask...upon reading ICR,
1840	 * interrupts are masked.  No need for the
1841	 * IMC write
1842	 */
1843
1844	if (icr & E1000_ICR_LSC) {
1845		hw->mac.get_link_status = true;
1846		/* ICH8 workaround-- Call gig speed drop workaround on cable
1847		 * disconnect (LSC) before accessing any PHY registers
1848		 */
1849		if ((adapter->flags & FLAG_LSC_GIG_SPEED_DROP) &&
1850		    (!(er32(STATUS) & E1000_STATUS_LU)))
1851			schedule_work(&adapter->downshift_task);
1852
1853		/* 80003ES2LAN workaround--
1854		 * For packet buffer work-around on link down event;
1855		 * disable receives here in the ISR and
1856		 * reset adapter in watchdog
1857		 */
1858		if (netif_carrier_ok(netdev) &&
1859		    (adapter->flags & FLAG_RX_NEEDS_RESTART)) {
1860			/* disable receives */
1861			rctl = er32(RCTL);
1862			ew32(RCTL, rctl & ~E1000_RCTL_EN);
1863			adapter->flags |= FLAG_RESTART_NOW;
1864		}
1865		/* guard against interrupt when we're going down */
1866		if (!test_bit(__E1000_DOWN, &adapter->state))
1867			mod_timer(&adapter->watchdog_timer, jiffies + 1);
1868	}
1869
1870	/* Reset on uncorrectable ECC error */
1871	if ((icr & E1000_ICR_ECCER) && (hw->mac.type >= e1000_pch_lpt)) {
1872		u32 pbeccsts = er32(PBECCSTS);
1873
1874		adapter->corr_errors +=
1875		    pbeccsts & E1000_PBECCSTS_CORR_ERR_CNT_MASK;
1876		adapter->uncorr_errors +=
1877		    (pbeccsts & E1000_PBECCSTS_UNCORR_ERR_CNT_MASK) >>
1878		    E1000_PBECCSTS_UNCORR_ERR_CNT_SHIFT;
1879
1880		/* Do the reset outside of interrupt context */
1881		schedule_work(&adapter->reset_task);
1882
1883		/* return immediately since reset is imminent */
1884		return IRQ_HANDLED;
1885	}
1886
1887	if (napi_schedule_prep(&adapter->napi)) {
1888		adapter->total_tx_bytes = 0;
1889		adapter->total_tx_packets = 0;
1890		adapter->total_rx_bytes = 0;
1891		adapter->total_rx_packets = 0;
1892		__napi_schedule(&adapter->napi);
1893	}
1894
1895	return IRQ_HANDLED;
1896}
1897
1898static irqreturn_t e1000_msix_other(int __always_unused irq, void *data)
1899{
1900	struct net_device *netdev = data;
1901	struct e1000_adapter *adapter = netdev_priv(netdev);
1902	struct e1000_hw *hw = &adapter->hw;
1903	u32 icr = er32(ICR);
1904
1905	if (icr & adapter->eiac_mask)
1906		ew32(ICS, (icr & adapter->eiac_mask));
1907
1908	if (icr & E1000_ICR_LSC) {
1909		hw->mac.get_link_status = true;
1910		/* guard against interrupt when we're going down */
1911		if (!test_bit(__E1000_DOWN, &adapter->state))
1912			mod_timer(&adapter->watchdog_timer, jiffies + 1);
1913	}
1914
1915	if (!test_bit(__E1000_DOWN, &adapter->state))
1916		ew32(IMS, E1000_IMS_OTHER | IMS_OTHER_MASK);
1917
1918	return IRQ_HANDLED;
1919}
1920
1921static irqreturn_t e1000_intr_msix_tx(int __always_unused irq, void *data)
1922{
1923	struct net_device *netdev = data;
1924	struct e1000_adapter *adapter = netdev_priv(netdev);
1925	struct e1000_hw *hw = &adapter->hw;
1926	struct e1000_ring *tx_ring = adapter->tx_ring;
1927
1928	adapter->total_tx_bytes = 0;
1929	adapter->total_tx_packets = 0;
1930
1931	if (!e1000_clean_tx_irq(tx_ring))
1932		/* Ring was not completely cleaned, so fire another interrupt */
1933		ew32(ICS, tx_ring->ims_val);
1934
1935	if (!test_bit(__E1000_DOWN, &adapter->state))
1936		ew32(IMS, adapter->tx_ring->ims_val);
1937
1938	return IRQ_HANDLED;
1939}
1940
1941static irqreturn_t e1000_intr_msix_rx(int __always_unused irq, void *data)
1942{
1943	struct net_device *netdev = data;
1944	struct e1000_adapter *adapter = netdev_priv(netdev);
1945	struct e1000_ring *rx_ring = adapter->rx_ring;
1946
1947	/* Write the ITR value calculated at the end of the
1948	 * previous interrupt.
1949	 */
1950	if (rx_ring->set_itr) {
1951		u32 itr = rx_ring->itr_val ?
1952			  1000000000 / (rx_ring->itr_val * 256) : 0;
1953
1954		writel(itr, rx_ring->itr_register);
1955		rx_ring->set_itr = 0;
1956	}
1957
1958	if (napi_schedule_prep(&adapter->napi)) {
1959		adapter->total_rx_bytes = 0;
1960		adapter->total_rx_packets = 0;
1961		__napi_schedule(&adapter->napi);
1962	}
1963	return IRQ_HANDLED;
1964}
1965
1966/**
1967 * e1000_configure_msix - Configure MSI-X hardware
1968 * @adapter: board private structure
1969 *
1970 * e1000_configure_msix sets up the hardware to properly
1971 * generate MSI-X interrupts.
1972 **/
1973static void e1000_configure_msix(struct e1000_adapter *adapter)
1974{
1975	struct e1000_hw *hw = &adapter->hw;
1976	struct e1000_ring *rx_ring = adapter->rx_ring;
1977	struct e1000_ring *tx_ring = adapter->tx_ring;
1978	int vector = 0;
1979	u32 ctrl_ext, ivar = 0;
1980
1981	adapter->eiac_mask = 0;
1982
1983	/* Workaround issue with spurious interrupts on 82574 in MSI-X mode */
1984	if (hw->mac.type == e1000_82574) {
1985		u32 rfctl = er32(RFCTL);
1986
1987		rfctl |= E1000_RFCTL_ACK_DIS;
1988		ew32(RFCTL, rfctl);
1989	}
1990
1991	/* Configure Rx vector */
1992	rx_ring->ims_val = E1000_IMS_RXQ0;
1993	adapter->eiac_mask |= rx_ring->ims_val;
1994	if (rx_ring->itr_val)
1995		writel(1000000000 / (rx_ring->itr_val * 256),
1996		       rx_ring->itr_register);
1997	else
1998		writel(1, rx_ring->itr_register);
1999	ivar = E1000_IVAR_INT_ALLOC_VALID | vector;
2000
2001	/* Configure Tx vector */
2002	tx_ring->ims_val = E1000_IMS_TXQ0;
2003	vector++;
2004	if (tx_ring->itr_val)
2005		writel(1000000000 / (tx_ring->itr_val * 256),
2006		       tx_ring->itr_register);
2007	else
2008		writel(1, tx_ring->itr_register);
2009	adapter->eiac_mask |= tx_ring->ims_val;
2010	ivar |= ((E1000_IVAR_INT_ALLOC_VALID | vector) << 8);
2011
2012	/* set vector for Other Causes, e.g. link changes */
2013	vector++;
2014	ivar |= ((E1000_IVAR_INT_ALLOC_VALID | vector) << 16);
2015	if (rx_ring->itr_val)
2016		writel(1000000000 / (rx_ring->itr_val * 256),
2017		       hw->hw_addr + E1000_EITR_82574(vector));
2018	else
2019		writel(1, hw->hw_addr + E1000_EITR_82574(vector));
2020
2021	/* Cause Tx interrupts on every write back */
2022	ivar |= BIT(31);
2023
2024	ew32(IVAR, ivar);
2025
2026	/* enable MSI-X PBA support */
2027	ctrl_ext = er32(CTRL_EXT) & ~E1000_CTRL_EXT_IAME;
2028	ctrl_ext |= E1000_CTRL_EXT_PBA_CLR | E1000_CTRL_EXT_EIAME;
2029	ew32(CTRL_EXT, ctrl_ext);
2030	e1e_flush();
2031}
2032
2033void e1000e_reset_interrupt_capability(struct e1000_adapter *adapter)
2034{
2035	if (adapter->msix_entries) {
2036		pci_disable_msix(adapter->pdev);
2037		kfree(adapter->msix_entries);
2038		adapter->msix_entries = NULL;
2039	} else if (adapter->flags & FLAG_MSI_ENABLED) {
2040		pci_disable_msi(adapter->pdev);
2041		adapter->flags &= ~FLAG_MSI_ENABLED;
2042	}
2043}
2044
2045/**
2046 * e1000e_set_interrupt_capability - set MSI or MSI-X if supported
2047 * @adapter: board private structure
2048 *
2049 * Attempt to configure interrupts using the best available
2050 * capabilities of the hardware and kernel.
2051 **/
2052void e1000e_set_interrupt_capability(struct e1000_adapter *adapter)
2053{
2054	int err;
2055	int i;
2056
2057	switch (adapter->int_mode) {
2058	case E1000E_INT_MODE_MSIX:
2059		if (adapter->flags & FLAG_HAS_MSIX) {
2060			adapter->num_vectors = 3; /* RxQ0, TxQ0 and other */
2061			adapter->msix_entries = kcalloc(adapter->num_vectors,
2062							sizeof(struct
2063							       msix_entry),
2064							GFP_KERNEL);
2065			if (adapter->msix_entries) {
2066				struct e1000_adapter *a = adapter;
2067
2068				for (i = 0; i < adapter->num_vectors; i++)
2069					adapter->msix_entries[i].entry = i;
2070
2071				err = pci_enable_msix_range(a->pdev,
2072							    a->msix_entries,
2073							    a->num_vectors,
2074							    a->num_vectors);
2075				if (err > 0)
2076					return;
2077			}
2078			/* MSI-X failed, so fall through and try MSI */
2079			e_err("Failed to initialize MSI-X interrupts.  Falling back to MSI interrupts.\n");
2080			e1000e_reset_interrupt_capability(adapter);
2081		}
2082		adapter->int_mode = E1000E_INT_MODE_MSI;
2083		fallthrough;
2084	case E1000E_INT_MODE_MSI:
2085		if (!pci_enable_msi(adapter->pdev)) {
2086			adapter->flags |= FLAG_MSI_ENABLED;
2087		} else {
2088			adapter->int_mode = E1000E_INT_MODE_LEGACY;
2089			e_err("Failed to initialize MSI interrupts.  Falling back to legacy interrupts.\n");
2090		}
2091		fallthrough;
2092	case E1000E_INT_MODE_LEGACY:
2093		/* Don't do anything; this is the system default */
2094		break;
2095	}
2096
2097	/* store the number of vectors being used */
2098	adapter->num_vectors = 1;
2099}
2100
2101/**
2102 * e1000_request_msix - Initialize MSI-X interrupts
2103 * @adapter: board private structure
2104 *
2105 * e1000_request_msix allocates MSI-X vectors and requests interrupts from the
2106 * kernel.
2107 **/
2108static int e1000_request_msix(struct e1000_adapter *adapter)
2109{
2110	struct net_device *netdev = adapter->netdev;
2111	int err = 0, vector = 0;
2112
2113	if (strlen(netdev->name) < (IFNAMSIZ - 5))
2114		snprintf(adapter->rx_ring->name,
2115			 sizeof(adapter->rx_ring->name) - 1,
2116			 "%.14s-rx-0", netdev->name);
2117	else
2118		memcpy(adapter->rx_ring->name, netdev->name, IFNAMSIZ);
2119	err = request_irq(adapter->msix_entries[vector].vector,
2120			  e1000_intr_msix_rx, 0, adapter->rx_ring->name,
2121			  netdev);
2122	if (err)
2123		return err;
2124	adapter->rx_ring->itr_register = adapter->hw.hw_addr +
2125	    E1000_EITR_82574(vector);
2126	adapter->rx_ring->itr_val = adapter->itr;
2127	vector++;
2128
2129	if (strlen(netdev->name) < (IFNAMSIZ - 5))
2130		snprintf(adapter->tx_ring->name,
2131			 sizeof(adapter->tx_ring->name) - 1,
2132			 "%.14s-tx-0", netdev->name);
2133	else
2134		memcpy(adapter->tx_ring->name, netdev->name, IFNAMSIZ);
2135	err = request_irq(adapter->msix_entries[vector].vector,
2136			  e1000_intr_msix_tx, 0, adapter->tx_ring->name,
2137			  netdev);
2138	if (err)
2139		return err;
2140	adapter->tx_ring->itr_register = adapter->hw.hw_addr +
2141	    E1000_EITR_82574(vector);
2142	adapter->tx_ring->itr_val = adapter->itr;
2143	vector++;
2144
2145	err = request_irq(adapter->msix_entries[vector].vector,
2146			  e1000_msix_other, 0, netdev->name, netdev);
2147	if (err)
2148		return err;
2149
2150	e1000_configure_msix(adapter);
2151
2152	return 0;
2153}
2154
2155/**
2156 * e1000_request_irq - initialize interrupts
2157 * @adapter: board private structure
2158 *
2159 * Attempts to configure interrupts using the best available
2160 * capabilities of the hardware and kernel.
2161 **/
2162static int e1000_request_irq(struct e1000_adapter *adapter)
2163{
2164	struct net_device *netdev = adapter->netdev;
2165	int err;
2166
2167	if (adapter->msix_entries) {
2168		err = e1000_request_msix(adapter);
2169		if (!err)
2170			return err;
2171		/* fall back to MSI */
2172		e1000e_reset_interrupt_capability(adapter);
2173		adapter->int_mode = E1000E_INT_MODE_MSI;
2174		e1000e_set_interrupt_capability(adapter);
2175	}
2176	if (adapter->flags & FLAG_MSI_ENABLED) {
2177		err = request_irq(adapter->pdev->irq, e1000_intr_msi, 0,
2178				  netdev->name, netdev);
2179		if (!err)
2180			return err;
2181
2182		/* fall back to legacy interrupt */
2183		e1000e_reset_interrupt_capability(adapter);
2184		adapter->int_mode = E1000E_INT_MODE_LEGACY;
2185	}
2186
2187	err = request_irq(adapter->pdev->irq, e1000_intr, IRQF_SHARED,
2188			  netdev->name, netdev);
2189	if (err)
2190		e_err("Unable to allocate interrupt, Error: %d\n", err);
2191
2192	return err;
2193}
2194
2195static void e1000_free_irq(struct e1000_adapter *adapter)
2196{
2197	struct net_device *netdev = adapter->netdev;
2198
2199	if (adapter->msix_entries) {
2200		int vector = 0;
2201
2202		free_irq(adapter->msix_entries[vector].vector, netdev);
2203		vector++;
2204
2205		free_irq(adapter->msix_entries[vector].vector, netdev);
2206		vector++;
2207
2208		/* Other Causes interrupt vector */
2209		free_irq(adapter->msix_entries[vector].vector, netdev);
2210		return;
2211	}
2212
2213	free_irq(adapter->pdev->irq, netdev);
2214}
2215
2216/**
2217 * e1000_irq_disable - Mask off interrupt generation on the NIC
2218 * @adapter: board private structure
2219 **/
2220static void e1000_irq_disable(struct e1000_adapter *adapter)
2221{
2222	struct e1000_hw *hw = &adapter->hw;
2223
2224	ew32(IMC, ~0);
2225	if (adapter->msix_entries)
2226		ew32(EIAC_82574, 0);
2227	e1e_flush();
2228
2229	if (adapter->msix_entries) {
2230		int i;
2231
2232		for (i = 0; i < adapter->num_vectors; i++)
2233			synchronize_irq(adapter->msix_entries[i].vector);
2234	} else {
2235		synchronize_irq(adapter->pdev->irq);
2236	}
2237}
2238
2239/**
2240 * e1000_irq_enable - Enable default interrupt generation settings
2241 * @adapter: board private structure
2242 **/
2243static void e1000_irq_enable(struct e1000_adapter *adapter)
2244{
2245	struct e1000_hw *hw = &adapter->hw;
2246
2247	if (adapter->msix_entries) {
2248		ew32(EIAC_82574, adapter->eiac_mask & E1000_EIAC_MASK_82574);
2249		ew32(IMS, adapter->eiac_mask | E1000_IMS_OTHER |
2250		     IMS_OTHER_MASK);
2251	} else if (hw->mac.type >= e1000_pch_lpt) {
2252		ew32(IMS, IMS_ENABLE_MASK | E1000_IMS_ECCER);
2253	} else {
2254		ew32(IMS, IMS_ENABLE_MASK);
2255	}
2256	e1e_flush();
2257}
2258
2259/**
2260 * e1000e_get_hw_control - get control of the h/w from f/w
2261 * @adapter: address of board private structure
2262 *
2263 * e1000e_get_hw_control sets {CTRL_EXT|SWSM}:DRV_LOAD bit.
2264 * For ASF and Pass Through versions of f/w this means that
2265 * the driver is loaded. For AMT version (only with 82573)
2266 * of the f/w this means that the network i/f is open.
2267 **/
2268void e1000e_get_hw_control(struct e1000_adapter *adapter)
2269{
2270	struct e1000_hw *hw = &adapter->hw;
2271	u32 ctrl_ext;
2272	u32 swsm;
2273
2274	/* Let firmware know the driver has taken over */
2275	if (adapter->flags & FLAG_HAS_SWSM_ON_LOAD) {
2276		swsm = er32(SWSM);
2277		ew32(SWSM, swsm | E1000_SWSM_DRV_LOAD);
2278	} else if (adapter->flags & FLAG_HAS_CTRLEXT_ON_LOAD) {
2279		ctrl_ext = er32(CTRL_EXT);
2280		ew32(CTRL_EXT, ctrl_ext | E1000_CTRL_EXT_DRV_LOAD);
2281	}
2282}
2283
2284/**
2285 * e1000e_release_hw_control - release control of the h/w to f/w
2286 * @adapter: address of board private structure
2287 *
2288 * e1000e_release_hw_control resets {CTRL_EXT|SWSM}:DRV_LOAD bit.
2289 * For ASF and Pass Through versions of f/w this means that the
2290 * driver is no longer loaded. For AMT version (only with 82573) i
2291 * of the f/w this means that the network i/f is closed.
2292 *
2293 **/
2294void e1000e_release_hw_control(struct e1000_adapter *adapter)
2295{
2296	struct e1000_hw *hw = &adapter->hw;
2297	u32 ctrl_ext;
2298	u32 swsm;
2299
2300	/* Let firmware taken over control of h/w */
2301	if (adapter->flags & FLAG_HAS_SWSM_ON_LOAD) {
2302		swsm = er32(SWSM);
2303		ew32(SWSM, swsm & ~E1000_SWSM_DRV_LOAD);
2304	} else if (adapter->flags & FLAG_HAS_CTRLEXT_ON_LOAD) {
2305		ctrl_ext = er32(CTRL_EXT);
2306		ew32(CTRL_EXT, ctrl_ext & ~E1000_CTRL_EXT_DRV_LOAD);
2307	}
2308}
2309
2310/**
2311 * e1000_alloc_ring_dma - allocate memory for a ring structure
2312 * @adapter: board private structure
2313 * @ring: ring struct for which to allocate dma
2314 **/
2315static int e1000_alloc_ring_dma(struct e1000_adapter *adapter,
2316				struct e1000_ring *ring)
2317{
2318	struct pci_dev *pdev = adapter->pdev;
2319
2320	ring->desc = dma_alloc_coherent(&pdev->dev, ring->size, &ring->dma,
2321					GFP_KERNEL);
2322	if (!ring->desc)
2323		return -ENOMEM;
2324
2325	return 0;
2326}
2327
2328/**
2329 * e1000e_setup_tx_resources - allocate Tx resources (Descriptors)
2330 * @tx_ring: Tx descriptor ring
2331 *
2332 * Return 0 on success, negative on failure
2333 **/
2334int e1000e_setup_tx_resources(struct e1000_ring *tx_ring)
2335{
2336	struct e1000_adapter *adapter = tx_ring->adapter;
2337	int err = -ENOMEM, size;
2338
2339	size = sizeof(struct e1000_buffer) * tx_ring->count;
2340	tx_ring->buffer_info = vzalloc(size);
2341	if (!tx_ring->buffer_info)
2342		goto err;
2343
2344	/* round up to nearest 4K */
2345	tx_ring->size = tx_ring->count * sizeof(struct e1000_tx_desc);
2346	tx_ring->size = ALIGN(tx_ring->size, 4096);
2347
2348	err = e1000_alloc_ring_dma(adapter, tx_ring);
2349	if (err)
2350		goto err;
2351
2352	tx_ring->next_to_use = 0;
2353	tx_ring->next_to_clean = 0;
2354
2355	return 0;
2356err:
2357	vfree(tx_ring->buffer_info);
2358	e_err("Unable to allocate memory for the transmit descriptor ring\n");
2359	return err;
2360}
2361
2362/**
2363 * e1000e_setup_rx_resources - allocate Rx resources (Descriptors)
2364 * @rx_ring: Rx descriptor ring
2365 *
2366 * Returns 0 on success, negative on failure
2367 **/
2368int e1000e_setup_rx_resources(struct e1000_ring *rx_ring)
2369{
2370	struct e1000_adapter *adapter = rx_ring->adapter;
2371	struct e1000_buffer *buffer_info;
2372	int i, size, desc_len, err = -ENOMEM;
2373
2374	size = sizeof(struct e1000_buffer) * rx_ring->count;
2375	rx_ring->buffer_info = vzalloc(size);
2376	if (!rx_ring->buffer_info)
2377		goto err;
2378
2379	for (i = 0; i < rx_ring->count; i++) {
2380		buffer_info = &rx_ring->buffer_info[i];
2381		buffer_info->ps_pages = kcalloc(PS_PAGE_BUFFERS,
2382						sizeof(struct e1000_ps_page),
2383						GFP_KERNEL);
2384		if (!buffer_info->ps_pages)
2385			goto err_pages;
2386	}
2387
2388	desc_len = sizeof(union e1000_rx_desc_packet_split);
2389
2390	/* Round up to nearest 4K */
2391	rx_ring->size = rx_ring->count * desc_len;
2392	rx_ring->size = ALIGN(rx_ring->size, 4096);
2393
2394	err = e1000_alloc_ring_dma(adapter, rx_ring);
2395	if (err)
2396		goto err_pages;
2397
2398	rx_ring->next_to_clean = 0;
2399	rx_ring->next_to_use = 0;
2400	rx_ring->rx_skb_top = NULL;
2401
2402	return 0;
2403
2404err_pages:
2405	for (i = 0; i < rx_ring->count; i++) {
2406		buffer_info = &rx_ring->buffer_info[i];
2407		kfree(buffer_info->ps_pages);
2408	}
2409err:
2410	vfree(rx_ring->buffer_info);
2411	e_err("Unable to allocate memory for the receive descriptor ring\n");
2412	return err;
2413}
2414
2415/**
2416 * e1000_clean_tx_ring - Free Tx Buffers
2417 * @tx_ring: Tx descriptor ring
2418 **/
2419static void e1000_clean_tx_ring(struct e1000_ring *tx_ring)
2420{
2421	struct e1000_adapter *adapter = tx_ring->adapter;
2422	struct e1000_buffer *buffer_info;
2423	unsigned long size;
2424	unsigned int i;
2425
2426	for (i = 0; i < tx_ring->count; i++) {
2427		buffer_info = &tx_ring->buffer_info[i];
2428		e1000_put_txbuf(tx_ring, buffer_info, false);
2429	}
2430
2431	netdev_reset_queue(adapter->netdev);
2432	size = sizeof(struct e1000_buffer) * tx_ring->count;
2433	memset(tx_ring->buffer_info, 0, size);
2434
2435	memset(tx_ring->desc, 0, tx_ring->size);
2436
2437	tx_ring->next_to_use = 0;
2438	tx_ring->next_to_clean = 0;
2439}
2440
2441/**
2442 * e1000e_free_tx_resources - Free Tx Resources per Queue
2443 * @tx_ring: Tx descriptor ring
2444 *
2445 * Free all transmit software resources
2446 **/
2447void e1000e_free_tx_resources(struct e1000_ring *tx_ring)
2448{
2449	struct e1000_adapter *adapter = tx_ring->adapter;
2450	struct pci_dev *pdev = adapter->pdev;
2451
2452	e1000_clean_tx_ring(tx_ring);
2453
2454	vfree(tx_ring->buffer_info);
2455	tx_ring->buffer_info = NULL;
2456
2457	dma_free_coherent(&pdev->dev, tx_ring->size, tx_ring->desc,
2458			  tx_ring->dma);
2459	tx_ring->desc = NULL;
2460}
2461
2462/**
2463 * e1000e_free_rx_resources - Free Rx Resources
2464 * @rx_ring: Rx descriptor ring
2465 *
2466 * Free all receive software resources
2467 **/
2468void e1000e_free_rx_resources(struct e1000_ring *rx_ring)
2469{
2470	struct e1000_adapter *adapter = rx_ring->adapter;
2471	struct pci_dev *pdev = adapter->pdev;
2472	int i;
2473
2474	e1000_clean_rx_ring(rx_ring);
2475
2476	for (i = 0; i < rx_ring->count; i++)
2477		kfree(rx_ring->buffer_info[i].ps_pages);
2478
2479	vfree(rx_ring->buffer_info);
2480	rx_ring->buffer_info = NULL;
2481
2482	dma_free_coherent(&pdev->dev, rx_ring->size, rx_ring->desc,
2483			  rx_ring->dma);
2484	rx_ring->desc = NULL;
2485}
2486
2487/**
2488 * e1000_update_itr - update the dynamic ITR value based on statistics
2489 * @itr_setting: current adapter->itr
2490 * @packets: the number of packets during this measurement interval
2491 * @bytes: the number of bytes during this measurement interval
2492 *
2493 *      Stores a new ITR value based on packets and byte
2494 *      counts during the last interrupt.  The advantage of per interrupt
2495 *      computation is faster updates and more accurate ITR for the current
2496 *      traffic pattern.  Constants in this function were computed
2497 *      based on theoretical maximum wire speed and thresholds were set based
2498 *      on testing data as well as attempting to minimize response time
2499 *      while increasing bulk throughput.  This functionality is controlled
2500 *      by the InterruptThrottleRate module parameter.
2501 **/
2502static unsigned int e1000_update_itr(u16 itr_setting, int packets, int bytes)
2503{
2504	unsigned int retval = itr_setting;
2505
2506	if (packets == 0)
2507		return itr_setting;
2508
2509	switch (itr_setting) {
2510	case lowest_latency:
2511		/* handle TSO and jumbo frames */
2512		if (bytes / packets > 8000)
2513			retval = bulk_latency;
2514		else if ((packets < 5) && (bytes > 512))
2515			retval = low_latency;
2516		break;
2517	case low_latency:	/* 50 usec aka 20000 ints/s */
2518		if (bytes > 10000) {
2519			/* this if handles the TSO accounting */
2520			if (bytes / packets > 8000)
2521				retval = bulk_latency;
2522			else if ((packets < 10) || ((bytes / packets) > 1200))
2523				retval = bulk_latency;
2524			else if ((packets > 35))
2525				retval = lowest_latency;
2526		} else if (bytes / packets > 2000) {
2527			retval = bulk_latency;
2528		} else if (packets <= 2 && bytes < 512) {
2529			retval = lowest_latency;
2530		}
2531		break;
2532	case bulk_latency:	/* 250 usec aka 4000 ints/s */
2533		if (bytes > 25000) {
2534			if (packets > 35)
2535				retval = low_latency;
2536		} else if (bytes < 6000) {
2537			retval = low_latency;
2538		}
2539		break;
2540	}
2541
2542	return retval;
2543}
2544
2545static void e1000_set_itr(struct e1000_adapter *adapter)
2546{
2547	u16 current_itr;
2548	u32 new_itr = adapter->itr;
2549
2550	/* for non-gigabit speeds, just fix the interrupt rate at 4000 */
2551	if (adapter->link_speed != SPEED_1000) {
2552		current_itr = 0;
2553		new_itr = 4000;
2554		goto set_itr_now;
2555	}
2556
2557	if (adapter->flags2 & FLAG2_DISABLE_AIM) {
2558		new_itr = 0;
2559		goto set_itr_now;
2560	}
2561
2562	adapter->tx_itr = e1000_update_itr(adapter->tx_itr,
2563					   adapter->total_tx_packets,
2564					   adapter->total_tx_bytes);
2565	/* conservative mode (itr 3) eliminates the lowest_latency setting */
2566	if (adapter->itr_setting == 3 && adapter->tx_itr == lowest_latency)
2567		adapter->tx_itr = low_latency;
2568
2569	adapter->rx_itr = e1000_update_itr(adapter->rx_itr,
2570					   adapter->total_rx_packets,
2571					   adapter->total_rx_bytes);
2572	/* conservative mode (itr 3) eliminates the lowest_latency setting */
2573	if (adapter->itr_setting == 3 && adapter->rx_itr == lowest_latency)
2574		adapter->rx_itr = low_latency;
2575
2576	current_itr = max(adapter->rx_itr, adapter->tx_itr);
2577
2578	/* counts and packets in update_itr are dependent on these numbers */
2579	switch (current_itr) {
2580	case lowest_latency:
2581		new_itr = 70000;
2582		break;
2583	case low_latency:
2584		new_itr = 20000;	/* aka hwitr = ~200 */
2585		break;
2586	case bulk_latency:
2587		new_itr = 4000;
2588		break;
2589	default:
2590		break;
2591	}
2592
2593set_itr_now:
2594	if (new_itr != adapter->itr) {
2595		/* this attempts to bias the interrupt rate towards Bulk
2596		 * by adding intermediate steps when interrupt rate is
2597		 * increasing
2598		 */
2599		new_itr = new_itr > adapter->itr ?
2600		    min(adapter->itr + (new_itr >> 2), new_itr) : new_itr;
2601		adapter->itr = new_itr;
2602		adapter->rx_ring->itr_val = new_itr;
2603		if (adapter->msix_entries)
2604			adapter->rx_ring->set_itr = 1;
2605		else
2606			e1000e_write_itr(adapter, new_itr);
2607	}
2608}
2609
2610/**
2611 * e1000e_write_itr - write the ITR value to the appropriate registers
2612 * @adapter: address of board private structure
2613 * @itr: new ITR value to program
2614 *
2615 * e1000e_write_itr determines if the adapter is in MSI-X mode
2616 * and, if so, writes the EITR registers with the ITR value.
2617 * Otherwise, it writes the ITR value into the ITR register.
2618 **/
2619void e1000e_write_itr(struct e1000_adapter *adapter, u32 itr)
2620{
2621	struct e1000_hw *hw = &adapter->hw;
2622	u32 new_itr = itr ? 1000000000 / (itr * 256) : 0;
2623
2624	if (adapter->msix_entries) {
2625		int vector;
2626
2627		for (vector = 0; vector < adapter->num_vectors; vector++)
2628			writel(new_itr, hw->hw_addr + E1000_EITR_82574(vector));
2629	} else {
2630		ew32(ITR, new_itr);
2631	}
2632}
2633
2634/**
2635 * e1000_alloc_queues - Allocate memory for all rings
2636 * @adapter: board private structure to initialize
2637 **/
2638static int e1000_alloc_queues(struct e1000_adapter *adapter)
2639{
2640	int size = sizeof(struct e1000_ring);
2641
2642	adapter->tx_ring = kzalloc(size, GFP_KERNEL);
2643	if (!adapter->tx_ring)
2644		goto err;
2645	adapter->tx_ring->count = adapter->tx_ring_count;
2646	adapter->tx_ring->adapter = adapter;
2647
2648	adapter->rx_ring = kzalloc(size, GFP_KERNEL);
2649	if (!adapter->rx_ring)
2650		goto err;
2651	adapter->rx_ring->count = adapter->rx_ring_count;
2652	adapter->rx_ring->adapter = adapter;
2653
2654	return 0;
2655err:
2656	e_err("Unable to allocate memory for queues\n");
2657	kfree(adapter->rx_ring);
2658	kfree(adapter->tx_ring);
2659	return -ENOMEM;
2660}
2661
2662/**
2663 * e1000e_poll - NAPI Rx polling callback
2664 * @napi: struct associated with this polling callback
2665 * @budget: number of packets driver is allowed to process this poll
2666 **/
2667static int e1000e_poll(struct napi_struct *napi, int budget)
2668{
2669	struct e1000_adapter *adapter = container_of(napi, struct e1000_adapter,
2670						     napi);
2671	struct e1000_hw *hw = &adapter->hw;
2672	struct net_device *poll_dev = adapter->netdev;
2673	int tx_cleaned = 1, work_done = 0;
2674
2675	adapter = netdev_priv(poll_dev);
2676
2677	if (!adapter->msix_entries ||
2678	    (adapter->rx_ring->ims_val & adapter->tx_ring->ims_val))
2679		tx_cleaned = e1000_clean_tx_irq(adapter->tx_ring);
2680
2681	adapter->clean_rx(adapter->rx_ring, &work_done, budget);
2682
2683	if (!tx_cleaned || work_done == budget)
2684		return budget;
2685
2686	/* Exit the polling mode, but don't re-enable interrupts if stack might
2687	 * poll us due to busy-polling
2688	 */
2689	if (likely(napi_complete_done(napi, work_done))) {
2690		if (adapter->itr_setting & 3)
2691			e1000_set_itr(adapter);
2692		if (!test_bit(__E1000_DOWN, &adapter->state)) {
2693			if (adapter->msix_entries)
2694				ew32(IMS, adapter->rx_ring->ims_val);
2695			else
2696				e1000_irq_enable(adapter);
2697		}
2698	}
2699
2700	return work_done;
2701}
2702
2703static int e1000_vlan_rx_add_vid(struct net_device *netdev,
2704				 __always_unused __be16 proto, u16 vid)
2705{
2706	struct e1000_adapter *adapter = netdev_priv(netdev);
2707	struct e1000_hw *hw = &adapter->hw;
2708	u32 vfta, index;
2709
2710	/* don't update vlan cookie if already programmed */
2711	if ((adapter->hw.mng_cookie.status &
2712	     E1000_MNG_DHCP_COOKIE_STATUS_VLAN) &&
2713	    (vid == adapter->mng_vlan_id))
2714		return 0;
2715
2716	/* add VID to filter table */
2717	if (adapter->flags & FLAG_HAS_HW_VLAN_FILTER) {
2718		index = (vid >> 5) & 0x7F;
2719		vfta = E1000_READ_REG_ARRAY(hw, E1000_VFTA, index);
2720		vfta |= BIT((vid & 0x1F));
2721		hw->mac.ops.write_vfta(hw, index, vfta);
2722	}
2723
2724	set_bit(vid, adapter->active_vlans);
2725
2726	return 0;
2727}
2728
2729static int e1000_vlan_rx_kill_vid(struct net_device *netdev,
2730				  __always_unused __be16 proto, u16 vid)
2731{
2732	struct e1000_adapter *adapter = netdev_priv(netdev);
2733	struct e1000_hw *hw = &adapter->hw;
2734	u32 vfta, index;
2735
2736	if ((adapter->hw.mng_cookie.status &
2737	     E1000_MNG_DHCP_COOKIE_STATUS_VLAN) &&
2738	    (vid == adapter->mng_vlan_id)) {
2739		/* release control to f/w */
2740		e1000e_release_hw_control(adapter);
2741		return 0;
2742	}
2743
2744	/* remove VID from filter table */
2745	if (adapter->flags & FLAG_HAS_HW_VLAN_FILTER) {
2746		index = (vid >> 5) & 0x7F;
2747		vfta = E1000_READ_REG_ARRAY(hw, E1000_VFTA, index);
2748		vfta &= ~BIT((vid & 0x1F));
2749		hw->mac.ops.write_vfta(hw, index, vfta);
2750	}
2751
2752	clear_bit(vid, adapter->active_vlans);
2753
2754	return 0;
2755}
2756
2757/**
2758 * e1000e_vlan_filter_disable - helper to disable hw VLAN filtering
2759 * @adapter: board private structure to initialize
2760 **/
2761static void e1000e_vlan_filter_disable(struct e1000_adapter *adapter)
2762{
2763	struct net_device *netdev = adapter->netdev;
2764	struct e1000_hw *hw = &adapter->hw;
2765	u32 rctl;
2766
2767	if (adapter->flags & FLAG_HAS_HW_VLAN_FILTER) {
2768		/* disable VLAN receive filtering */
2769		rctl = er32(RCTL);
2770		rctl &= ~(E1000_RCTL_VFE | E1000_RCTL_CFIEN);
2771		ew32(RCTL, rctl);
2772
2773		if (adapter->mng_vlan_id != (u16)E1000_MNG_VLAN_NONE) {
2774			e1000_vlan_rx_kill_vid(netdev, htons(ETH_P_8021Q),
2775					       adapter->mng_vlan_id);
2776			adapter->mng_vlan_id = E1000_MNG_VLAN_NONE;
2777		}
2778	}
2779}
2780
2781/**
2782 * e1000e_vlan_filter_enable - helper to enable HW VLAN filtering
2783 * @adapter: board private structure to initialize
2784 **/
2785static void e1000e_vlan_filter_enable(struct e1000_adapter *adapter)
2786{
2787	struct e1000_hw *hw = &adapter->hw;
2788	u32 rctl;
2789
2790	if (adapter->flags & FLAG_HAS_HW_VLAN_FILTER) {
2791		/* enable VLAN receive filtering */
2792		rctl = er32(RCTL);
2793		rctl |= E1000_RCTL_VFE;
2794		rctl &= ~E1000_RCTL_CFIEN;
2795		ew32(RCTL, rctl);
2796	}
2797}
2798
2799/**
2800 * e1000e_vlan_strip_disable - helper to disable HW VLAN stripping
2801 * @adapter: board private structure to initialize
2802 **/
2803static void e1000e_vlan_strip_disable(struct e1000_adapter *adapter)
2804{
2805	struct e1000_hw *hw = &adapter->hw;
2806	u32 ctrl;
2807
2808	/* disable VLAN tag insert/strip */
2809	ctrl = er32(CTRL);
2810	ctrl &= ~E1000_CTRL_VME;
2811	ew32(CTRL, ctrl);
2812}
2813
2814/**
2815 * e1000e_vlan_strip_enable - helper to enable HW VLAN stripping
2816 * @adapter: board private structure to initialize
2817 **/
2818static void e1000e_vlan_strip_enable(struct e1000_adapter *adapter)
2819{
2820	struct e1000_hw *hw = &adapter->hw;
2821	u32 ctrl;
2822
2823	/* enable VLAN tag insert/strip */
2824	ctrl = er32(CTRL);
2825	ctrl |= E1000_CTRL_VME;
2826	ew32(CTRL, ctrl);
2827}
2828
2829static void e1000_update_mng_vlan(struct e1000_adapter *adapter)
2830{
2831	struct net_device *netdev = adapter->netdev;
2832	u16 vid = adapter->hw.mng_cookie.vlan_id;
2833	u16 old_vid = adapter->mng_vlan_id;
2834
2835	if (adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN) {
2836		e1000_vlan_rx_add_vid(netdev, htons(ETH_P_8021Q), vid);
2837		adapter->mng_vlan_id = vid;
2838	}
2839
2840	if ((old_vid != (u16)E1000_MNG_VLAN_NONE) && (vid != old_vid))
2841		e1000_vlan_rx_kill_vid(netdev, htons(ETH_P_8021Q), old_vid);
2842}
2843
2844static void e1000_restore_vlan(struct e1000_adapter *adapter)
2845{
2846	u16 vid;
2847
2848	e1000_vlan_rx_add_vid(adapter->netdev, htons(ETH_P_8021Q), 0);
2849
2850	for_each_set_bit(vid, adapter->active_vlans, VLAN_N_VID)
2851	    e1000_vlan_rx_add_vid(adapter->netdev, htons(ETH_P_8021Q), vid);
2852}
2853
2854static void e1000_init_manageability_pt(struct e1000_adapter *adapter)
2855{
2856	struct e1000_hw *hw = &adapter->hw;
2857	u32 manc, manc2h, mdef, i, j;
2858
2859	if (!(adapter->flags & FLAG_MNG_PT_ENABLED))
2860		return;
2861
2862	manc = er32(MANC);
2863
2864	/* enable receiving management packets to the host. this will probably
2865	 * generate destination unreachable messages from the host OS, but
2866	 * the packets will be handled on SMBUS
2867	 */
2868	manc |= E1000_MANC_EN_MNG2HOST;
2869	manc2h = er32(MANC2H);
2870
2871	switch (hw->mac.type) {
2872	default:
2873		manc2h |= (E1000_MANC2H_PORT_623 | E1000_MANC2H_PORT_664);
2874		break;
2875	case e1000_82574:
2876	case e1000_82583:
2877		/* Check if IPMI pass-through decision filter already exists;
2878		 * if so, enable it.
2879		 */
2880		for (i = 0, j = 0; i < 8; i++) {
2881			mdef = er32(MDEF(i));
2882
2883			/* Ignore filters with anything other than IPMI ports */
2884			if (mdef & ~(E1000_MDEF_PORT_623 | E1000_MDEF_PORT_664))
2885				continue;
2886
2887			/* Enable this decision filter in MANC2H */
2888			if (mdef)
2889				manc2h |= BIT(i);
2890
2891			j |= mdef;
2892		}
2893
2894		if (j == (E1000_MDEF_PORT_623 | E1000_MDEF_PORT_664))
2895			break;
2896
2897		/* Create new decision filter in an empty filter */
2898		for (i = 0, j = 0; i < 8; i++)
2899			if (er32(MDEF(i)) == 0) {
2900				ew32(MDEF(i), (E1000_MDEF_PORT_623 |
2901					       E1000_MDEF_PORT_664));
2902				manc2h |= BIT(1);
2903				j++;
2904				break;
2905			}
2906
2907		if (!j)
2908			e_warn("Unable to create IPMI pass-through filter\n");
2909		break;
2910	}
2911
2912	ew32(MANC2H, manc2h);
2913	ew32(MANC, manc);
2914}
2915
2916/**
2917 * e1000_configure_tx - Configure Transmit Unit after Reset
2918 * @adapter: board private structure
2919 *
2920 * Configure the Tx unit of the MAC after a reset.
2921 **/
2922static void e1000_configure_tx(struct e1000_adapter *adapter)
2923{
2924	struct e1000_hw *hw = &adapter->hw;
2925	struct e1000_ring *tx_ring = adapter->tx_ring;
2926	u64 tdba;
2927	u32 tdlen, tctl, tarc;
2928
2929	/* Setup the HW Tx Head and Tail descriptor pointers */
2930	tdba = tx_ring->dma;
2931	tdlen = tx_ring->count * sizeof(struct e1000_tx_desc);
2932	ew32(TDBAL(0), (tdba & DMA_BIT_MASK(32)));
2933	ew32(TDBAH(0), (tdba >> 32));
2934	ew32(TDLEN(0), tdlen);
2935	ew32(TDH(0), 0);
2936	ew32(TDT(0), 0);
2937	tx_ring->head = adapter->hw.hw_addr + E1000_TDH(0);
2938	tx_ring->tail = adapter->hw.hw_addr + E1000_TDT(0);
2939
2940	writel(0, tx_ring->head);
2941	if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
2942		e1000e_update_tdt_wa(tx_ring, 0);
2943	else
2944		writel(0, tx_ring->tail);
2945
2946	/* Set the Tx Interrupt Delay register */
2947	ew32(TIDV, adapter->tx_int_delay);
2948	/* Tx irq moderation */
2949	ew32(TADV, adapter->tx_abs_int_delay);
2950
2951	if (adapter->flags2 & FLAG2_DMA_BURST) {
2952		u32 txdctl = er32(TXDCTL(0));
2953
2954		txdctl &= ~(E1000_TXDCTL_PTHRESH | E1000_TXDCTL_HTHRESH |
2955			    E1000_TXDCTL_WTHRESH);
2956		/* set up some performance related parameters to encourage the
2957		 * hardware to use the bus more efficiently in bursts, depends
2958		 * on the tx_int_delay to be enabled,
2959		 * wthresh = 1 ==> burst write is disabled to avoid Tx stalls
2960		 * hthresh = 1 ==> prefetch when one or more available
2961		 * pthresh = 0x1f ==> prefetch if internal cache 31 or less
2962		 * BEWARE: this seems to work but should be considered first if
2963		 * there are Tx hangs or other Tx related bugs
2964		 */
2965		txdctl |= E1000_TXDCTL_DMA_BURST_ENABLE;
2966		ew32(TXDCTL(0), txdctl);
2967	}
2968	/* erratum work around: set txdctl the same for both queues */
2969	ew32(TXDCTL(1), er32(TXDCTL(0)));
2970
2971	/* Program the Transmit Control Register */
2972	tctl = er32(TCTL);
2973	tctl &= ~E1000_TCTL_CT;
2974	tctl |= E1000_TCTL_PSP | E1000_TCTL_RTLC |
2975		(E1000_COLLISION_THRESHOLD << E1000_CT_SHIFT);
2976
2977	if (adapter->flags & FLAG_TARC_SPEED_MODE_BIT) {
2978		tarc = er32(TARC(0));
2979		/* set the speed mode bit, we'll clear it if we're not at
2980		 * gigabit link later
2981		 */
2982#define SPEED_MODE_BIT BIT(21)
2983		tarc |= SPEED_MODE_BIT;
2984		ew32(TARC(0), tarc);
2985	}
2986
2987	/* errata: program both queues to unweighted RR */
2988	if (adapter->flags & FLAG_TARC_SET_BIT_ZERO) {
2989		tarc = er32(TARC(0));
2990		tarc |= 1;
2991		ew32(TARC(0), tarc);
2992		tarc = er32(TARC(1));
2993		tarc |= 1;
2994		ew32(TARC(1), tarc);
2995	}
2996
2997	/* Setup Transmit Descriptor Settings for eop descriptor */
2998	adapter->txd_cmd = E1000_TXD_CMD_EOP | E1000_TXD_CMD_IFCS;
2999
3000	/* only set IDE if we are delaying interrupts using the timers */
3001	if (adapter->tx_int_delay)
3002		adapter->txd_cmd |= E1000_TXD_CMD_IDE;
3003
3004	/* enable Report Status bit */
3005	adapter->txd_cmd |= E1000_TXD_CMD_RS;
3006
3007	ew32(TCTL, tctl);
3008
3009	hw->mac.ops.config_collision_dist(hw);
3010
3011	/* SPT and KBL Si errata workaround to avoid data corruption */
3012	if (hw->mac.type == e1000_pch_spt) {
3013		u32 reg_val;
3014
3015		reg_val = er32(IOSFPC);
3016		reg_val |= E1000_RCTL_RDMTS_HEX;
3017		ew32(IOSFPC, reg_val);
3018
3019		reg_val = er32(TARC(0));
3020		/* SPT and KBL Si errata workaround to avoid Tx hang.
3021		 * Dropping the number of outstanding requests from
3022		 * 3 to 2 in order to avoid a buffer overrun.
3023		 */
3024		reg_val &= ~E1000_TARC0_CB_MULTIQ_3_REQ;
3025		reg_val |= E1000_TARC0_CB_MULTIQ_2_REQ;
3026		ew32(TARC(0), reg_val);
3027	}
3028}
3029
3030#define PAGE_USE_COUNT(S) (((S) >> PAGE_SHIFT) + \
3031			   (((S) & (PAGE_SIZE - 1)) ? 1 : 0))
3032
3033/**
3034 * e1000_setup_rctl - configure the receive control registers
3035 * @adapter: Board private structure
3036 **/
3037static void e1000_setup_rctl(struct e1000_adapter *adapter)
3038{
3039	struct e1000_hw *hw = &adapter->hw;
3040	u32 rctl, rfctl;
3041	u32 pages = 0;
3042
3043	/* Workaround Si errata on PCHx - configure jumbo frame flow.
3044	 * If jumbo frames not set, program related MAC/PHY registers
3045	 * to h/w defaults
3046	 */
3047	if (hw->mac.type >= e1000_pch2lan) {
3048		s32 ret_val;
3049
3050		if (adapter->netdev->mtu > ETH_DATA_LEN)
3051			ret_val = e1000_lv_jumbo_workaround_ich8lan(hw, true);
3052		else
3053			ret_val = e1000_lv_jumbo_workaround_ich8lan(hw, false);
3054
3055		if (ret_val)
3056			e_dbg("failed to enable|disable jumbo frame workaround mode\n");
3057	}
3058
3059	/* Program MC offset vector base */
3060	rctl = er32(RCTL);
3061	rctl &= ~(3 << E1000_RCTL_MO_SHIFT);
3062	rctl |= E1000_RCTL_EN | E1000_RCTL_BAM |
3063	    E1000_RCTL_LBM_NO | E1000_RCTL_RDMTS_HALF |
3064	    (adapter->hw.mac.mc_filter_type << E1000_RCTL_MO_SHIFT);
3065
3066	/* Do not Store bad packets */
3067	rctl &= ~E1000_RCTL_SBP;
3068
3069	/* Enable Long Packet receive */
3070	if (adapter->netdev->mtu <= ETH_DATA_LEN)
3071		rctl &= ~E1000_RCTL_LPE;
3072	else
3073		rctl |= E1000_RCTL_LPE;
3074
3075	/* Some systems expect that the CRC is included in SMBUS traffic. The
3076	 * hardware strips the CRC before sending to both SMBUS (BMC) and to
3077	 * host memory when this is enabled
3078	 */
3079	if (adapter->flags2 & FLAG2_CRC_STRIPPING)
3080		rctl |= E1000_RCTL_SECRC;
3081
3082	/* Workaround Si errata on 82577 PHY - configure IPG for jumbos */
3083	if ((hw->phy.type == e1000_phy_82577) && (rctl & E1000_RCTL_LPE)) {
3084		u16 phy_data;
3085
3086		e1e_rphy(hw, PHY_REG(770, 26), &phy_data);
3087		phy_data &= 0xfff8;
3088		phy_data |= BIT(2);
3089		e1e_wphy(hw, PHY_REG(770, 26), phy_data);
3090
3091		e1e_rphy(hw, 22, &phy_data);
3092		phy_data &= 0x0fff;
3093		phy_data |= BIT(14);
3094		e1e_wphy(hw, 0x10, 0x2823);
3095		e1e_wphy(hw, 0x11, 0x0003);
3096		e1e_wphy(hw, 22, phy_data);
3097	}
3098
3099	/* Setup buffer sizes */
3100	rctl &= ~E1000_RCTL_SZ_4096;
3101	rctl |= E1000_RCTL_BSEX;
3102	switch (adapter->rx_buffer_len) {
3103	case 2048:
3104	default:
3105		rctl |= E1000_RCTL_SZ_2048;
3106		rctl &= ~E1000_RCTL_BSEX;
3107		break;
3108	case 4096:
3109		rctl |= E1000_RCTL_SZ_4096;
3110		break;
3111	case 8192:
3112		rctl |= E1000_RCTL_SZ_8192;
3113		break;
3114	case 16384:
3115		rctl |= E1000_RCTL_SZ_16384;
3116		break;
3117	}
3118
3119	/* Enable Extended Status in all Receive Descriptors */
3120	rfctl = er32(RFCTL);
3121	rfctl |= E1000_RFCTL_EXTEN;
3122	ew32(RFCTL, rfctl);
3123
3124	/* 82571 and greater support packet-split where the protocol
3125	 * header is placed in skb->data and the packet data is
3126	 * placed in pages hanging off of skb_shinfo(skb)->nr_frags.
3127	 * In the case of a non-split, skb->data is linearly filled,
3128	 * followed by the page buffers.  Therefore, skb->data is
3129	 * sized to hold the largest protocol header.
3130	 *
3131	 * allocations using alloc_page take too long for regular MTU
3132	 * so only enable packet split for jumbo frames
3133	 *
3134	 * Using pages when the page size is greater than 16k wastes
3135	 * a lot of memory, since we allocate 3 pages at all times
3136	 * per packet.
3137	 */
3138	pages = PAGE_USE_COUNT(adapter->netdev->mtu);
3139	if ((pages <= 3) && (PAGE_SIZE <= 16384) && (rctl & E1000_RCTL_LPE))
3140		adapter->rx_ps_pages = pages;
3141	else
3142		adapter->rx_ps_pages = 0;
3143
3144	if (adapter->rx_ps_pages) {
3145		u32 psrctl = 0;
3146
3147		/* Enable Packet split descriptors */
3148		rctl |= E1000_RCTL_DTYP_PS;
3149
3150		psrctl |= adapter->rx_ps_bsize0 >> E1000_PSRCTL_BSIZE0_SHIFT;
3151
3152		switch (adapter->rx_ps_pages) {
3153		case 3:
3154			psrctl |= PAGE_SIZE << E1000_PSRCTL_BSIZE3_SHIFT;
3155			fallthrough;
3156		case 2:
3157			psrctl |= PAGE_SIZE << E1000_PSRCTL_BSIZE2_SHIFT;
3158			fallthrough;
3159		case 1:
3160			psrctl |= PAGE_SIZE >> E1000_PSRCTL_BSIZE1_SHIFT;
3161			break;
3162		}
3163
3164		ew32(PSRCTL, psrctl);
3165	}
3166
3167	/* This is useful for sniffing bad packets. */
3168	if (adapter->netdev->features & NETIF_F_RXALL) {
3169		/* UPE and MPE will be handled by normal PROMISC logic
3170		 * in e1000e_set_rx_mode
3171		 */
3172		rctl |= (E1000_RCTL_SBP |	/* Receive bad packets */
3173			 E1000_RCTL_BAM |	/* RX All Bcast Pkts */
3174			 E1000_RCTL_PMCF);	/* RX All MAC Ctrl Pkts */
3175
3176		rctl &= ~(E1000_RCTL_VFE |	/* Disable VLAN filter */
3177			  E1000_RCTL_DPF |	/* Allow filtered pause */
3178			  E1000_RCTL_CFIEN);	/* Dis VLAN CFIEN Filter */
3179		/* Do not mess with E1000_CTRL_VME, it affects transmit as well,
3180		 * and that breaks VLANs.
3181		 */
3182	}
3183
3184	ew32(RCTL, rctl);
3185	/* just started the receive unit, no need to restart */
3186	adapter->flags &= ~FLAG_RESTART_NOW;
3187}
3188
3189/**
3190 * e1000_configure_rx - Configure Receive Unit after Reset
3191 * @adapter: board private structure
3192 *
3193 * Configure the Rx unit of the MAC after a reset.
3194 **/
3195static void e1000_configure_rx(struct e1000_adapter *adapter)
3196{
3197	struct e1000_hw *hw = &adapter->hw;
3198	struct e1000_ring *rx_ring = adapter->rx_ring;
3199	u64 rdba;
3200	u32 rdlen, rctl, rxcsum, ctrl_ext;
3201
3202	if (adapter->rx_ps_pages) {
3203		/* this is a 32 byte descriptor */
3204		rdlen = rx_ring->count *
3205		    sizeof(union e1000_rx_desc_packet_split);
3206		adapter->clean_rx = e1000_clean_rx_irq_ps;
3207		adapter->alloc_rx_buf = e1000_alloc_rx_buffers_ps;
3208	} else if (adapter->netdev->mtu > ETH_FRAME_LEN + ETH_FCS_LEN) {
3209		rdlen = rx_ring->count * sizeof(union e1000_rx_desc_extended);
3210		adapter->clean_rx = e1000_clean_jumbo_rx_irq;
3211		adapter->alloc_rx_buf = e1000_alloc_jumbo_rx_buffers;
3212	} else {
3213		rdlen = rx_ring->count * sizeof(union e1000_rx_desc_extended);
3214		adapter->clean_rx = e1000_clean_rx_irq;
3215		adapter->alloc_rx_buf = e1000_alloc_rx_buffers;
3216	}
3217
3218	/* disable receives while setting up the descriptors */
3219	rctl = er32(RCTL);
3220	if (!(adapter->flags2 & FLAG2_NO_DISABLE_RX))
3221		ew32(RCTL, rctl & ~E1000_RCTL_EN);
3222	e1e_flush();
3223	usleep_range(10000, 11000);
3224
3225	if (adapter->flags2 & FLAG2_DMA_BURST) {
3226		/* set the writeback threshold (only takes effect if the RDTR
3227		 * is set). set GRAN=1 and write back up to 0x4 worth, and
3228		 * enable prefetching of 0x20 Rx descriptors
3229		 * granularity = 01
3230		 * wthresh = 04,
3231		 * hthresh = 04,
3232		 * pthresh = 0x20
3233		 */
3234		ew32(RXDCTL(0), E1000_RXDCTL_DMA_BURST_ENABLE);
3235		ew32(RXDCTL(1), E1000_RXDCTL_DMA_BURST_ENABLE);
3236	}
3237
3238	/* set the Receive Delay Timer Register */
3239	ew32(RDTR, adapter->rx_int_delay);
3240
3241	/* irq moderation */
3242	ew32(RADV, adapter->rx_abs_int_delay);
3243	if ((adapter->itr_setting != 0) && (adapter->itr != 0))
3244		e1000e_write_itr(adapter, adapter->itr);
3245
3246	ctrl_ext = er32(CTRL_EXT);
3247	/* Auto-Mask interrupts upon ICR access */
3248	ctrl_ext |= E1000_CTRL_EXT_IAME;
3249	ew32(IAM, 0xffffffff);
3250	ew32(CTRL_EXT, ctrl_ext);
3251	e1e_flush();
3252
3253	/* Setup the HW Rx Head and Tail Descriptor Pointers and
3254	 * the Base and Length of the Rx Descriptor Ring
3255	 */
3256	rdba = rx_ring->dma;
3257	ew32(RDBAL(0), (rdba & DMA_BIT_MASK(32)));
3258	ew32(RDBAH(0), (rdba >> 32));
3259	ew32(RDLEN(0), rdlen);
3260	ew32(RDH(0), 0);
3261	ew32(RDT(0), 0);
3262	rx_ring->head = adapter->hw.hw_addr + E1000_RDH(0);
3263	rx_ring->tail = adapter->hw.hw_addr + E1000_RDT(0);
3264
3265	writel(0, rx_ring->head);
3266	if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
3267		e1000e_update_rdt_wa(rx_ring, 0);
3268	else
3269		writel(0, rx_ring->tail);
3270
3271	/* Enable Receive Checksum Offload for TCP and UDP */
3272	rxcsum = er32(RXCSUM);
3273	if (adapter->netdev->features & NETIF_F_RXCSUM)
3274		rxcsum |= E1000_RXCSUM_TUOFL;
3275	else
3276		rxcsum &= ~E1000_RXCSUM_TUOFL;
3277	ew32(RXCSUM, rxcsum);
3278
3279	/* With jumbo frames, excessive C-state transition latencies result
3280	 * in dropped transactions.
3281	 */
3282	if (adapter->netdev->mtu > ETH_DATA_LEN) {
3283		u32 lat =
3284		    ((er32(PBA) & E1000_PBA_RXA_MASK) * 1024 -
3285		     adapter->max_frame_size) * 8 / 1000;
3286
3287		if (adapter->flags & FLAG_IS_ICH) {
3288			u32 rxdctl = er32(RXDCTL(0));
3289
3290			ew32(RXDCTL(0), rxdctl | 0x3 | BIT(8));
3291		}
3292
3293		dev_info(&adapter->pdev->dev,
3294			 "Some CPU C-states have been disabled in order to enable jumbo frames\n");
3295		cpu_latency_qos_update_request(&adapter->pm_qos_req, lat);
3296	} else {
3297		cpu_latency_qos_update_request(&adapter->pm_qos_req,
3298					       PM_QOS_DEFAULT_VALUE);
3299	}
3300
3301	/* Enable Receives */
3302	ew32(RCTL, rctl);
3303}
3304
3305/**
3306 * e1000e_write_mc_addr_list - write multicast addresses to MTA
3307 * @netdev: network interface device structure
3308 *
3309 * Writes multicast address list to the MTA hash table.
3310 * Returns: -ENOMEM on failure
3311 *                0 on no addresses written
3312 *                X on writing X addresses to MTA
3313 */
3314static int e1000e_write_mc_addr_list(struct net_device *netdev)
3315{
3316	struct e1000_adapter *adapter = netdev_priv(netdev);
3317	struct e1000_hw *hw = &adapter->hw;
3318	struct netdev_hw_addr *ha;
3319	u8 *mta_list;
3320	int i;
3321
3322	if (netdev_mc_empty(netdev)) {
3323		/* nothing to program, so clear mc list */
3324		hw->mac.ops.update_mc_addr_list(hw, NULL, 0);
3325		return 0;
3326	}
3327
3328	mta_list = kcalloc(netdev_mc_count(netdev), ETH_ALEN, GFP_ATOMIC);
3329	if (!mta_list)
3330		return -ENOMEM;
3331
3332	/* update_mc_addr_list expects a packed array of only addresses. */
3333	i = 0;
3334	netdev_for_each_mc_addr(ha, netdev)
3335	    memcpy(mta_list + (i++ * ETH_ALEN), ha->addr, ETH_ALEN);
3336
3337	hw->mac.ops.update_mc_addr_list(hw, mta_list, i);
3338	kfree(mta_list);
3339
3340	return netdev_mc_count(netdev);
3341}
3342
3343/**
3344 * e1000e_write_uc_addr_list - write unicast addresses to RAR table
3345 * @netdev: network interface device structure
3346 *
3347 * Writes unicast address list to the RAR table.
3348 * Returns: -ENOMEM on failure/insufficient address space
3349 *                0 on no addresses written
3350 *                X on writing X addresses to the RAR table
3351 **/
3352static int e1000e_write_uc_addr_list(struct net_device *netdev)
3353{
3354	struct e1000_adapter *adapter = netdev_priv(netdev);
3355	struct e1000_hw *hw = &adapter->hw;
3356	unsigned int rar_entries;
3357	int count = 0;
3358
3359	rar_entries = hw->mac.ops.rar_get_count(hw);
3360
3361	/* save a rar entry for our hardware address */
3362	rar_entries--;
3363
3364	/* save a rar entry for the LAA workaround */
3365	if (adapter->flags & FLAG_RESET_OVERWRITES_LAA)
3366		rar_entries--;
3367
3368	/* return ENOMEM indicating insufficient memory for addresses */
3369	if (netdev_uc_count(netdev) > rar_entries)
3370		return -ENOMEM;
3371
3372	if (!netdev_uc_empty(netdev) && rar_entries) {
3373		struct netdev_hw_addr *ha;
3374
3375		/* write the addresses in reverse order to avoid write
3376		 * combining
3377		 */
3378		netdev_for_each_uc_addr(ha, netdev) {
3379			int ret_val;
3380
3381			if (!rar_entries)
3382				break;
3383			ret_val = hw->mac.ops.rar_set(hw, ha->addr, rar_entries--);
3384			if (ret_val < 0)
3385				return -ENOMEM;
3386			count++;
3387		}
3388	}
3389
3390	/* zero out the remaining RAR entries not used above */
3391	for (; rar_entries > 0; rar_entries--) {
3392		ew32(RAH(rar_entries), 0);
3393		ew32(RAL(rar_entries), 0);
3394	}
3395	e1e_flush();
3396
3397	return count;
3398}
3399
3400/**
3401 * e1000e_set_rx_mode - secondary unicast, Multicast and Promiscuous mode set
3402 * @netdev: network interface device structure
3403 *
3404 * The ndo_set_rx_mode entry point is called whenever the unicast or multicast
3405 * address list or the network interface flags are updated.  This routine is
3406 * responsible for configuring the hardware for proper unicast, multicast,
3407 * promiscuous mode, and all-multi behavior.
3408 **/
3409static void e1000e_set_rx_mode(struct net_device *netdev)
3410{
3411	struct e1000_adapter *adapter = netdev_priv(netdev);
3412	struct e1000_hw *hw = &adapter->hw;
3413	u32 rctl;
3414
3415	if (pm_runtime_suspended(netdev->dev.parent))
3416		return;
3417
3418	/* Check for Promiscuous and All Multicast modes */
3419	rctl = er32(RCTL);
3420
3421	/* clear the affected bits */
3422	rctl &= ~(E1000_RCTL_UPE | E1000_RCTL_MPE);
3423
3424	if (netdev->flags & IFF_PROMISC) {
3425		rctl |= (E1000_RCTL_UPE | E1000_RCTL_MPE);
3426		/* Do not hardware filter VLANs in promisc mode */
3427		e1000e_vlan_filter_disable(adapter);
3428	} else {
3429		int count;
3430
3431		if (netdev->flags & IFF_ALLMULTI) {
3432			rctl |= E1000_RCTL_MPE;
3433		} else {
3434			/* Write addresses to the MTA, if the attempt fails
3435			 * then we should just turn on promiscuous mode so
3436			 * that we can at least receive multicast traffic
3437			 */
3438			count = e1000e_write_mc_addr_list(netdev);
3439			if (count < 0)
3440				rctl |= E1000_RCTL_MPE;
3441		}
3442		e1000e_vlan_filter_enable(adapter);
3443		/* Write addresses to available RAR registers, if there is not
3444		 * sufficient space to store all the addresses then enable
3445		 * unicast promiscuous mode
3446		 */
3447		count = e1000e_write_uc_addr_list(netdev);
3448		if (count < 0)
3449			rctl |= E1000_RCTL_UPE;
3450	}
3451
3452	ew32(RCTL, rctl);
3453
3454	if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX)
3455		e1000e_vlan_strip_enable(adapter);
3456	else
3457		e1000e_vlan_strip_disable(adapter);
3458}
3459
3460static void e1000e_setup_rss_hash(struct e1000_adapter *adapter)
3461{
3462	struct e1000_hw *hw = &adapter->hw;
3463	u32 mrqc, rxcsum;
3464	u32 rss_key[10];
3465	int i;
3466
3467	netdev_rss_key_fill(rss_key, sizeof(rss_key));
3468	for (i = 0; i < 10; i++)
3469		ew32(RSSRK(i), rss_key[i]);
3470
3471	/* Direct all traffic to queue 0 */
3472	for (i = 0; i < 32; i++)
3473		ew32(RETA(i), 0);
3474
3475	/* Disable raw packet checksumming so that RSS hash is placed in
3476	 * descriptor on writeback.
3477	 */
3478	rxcsum = er32(RXCSUM);
3479	rxcsum |= E1000_RXCSUM_PCSD;
3480
3481	ew32(RXCSUM, rxcsum);
3482
3483	mrqc = (E1000_MRQC_RSS_FIELD_IPV4 |
3484		E1000_MRQC_RSS_FIELD_IPV4_TCP |
3485		E1000_MRQC_RSS_FIELD_IPV6 |
3486		E1000_MRQC_RSS_FIELD_IPV6_TCP |
3487		E1000_MRQC_RSS_FIELD_IPV6_TCP_EX);
3488
3489	ew32(MRQC, mrqc);
3490}
3491
3492/**
3493 * e1000e_get_base_timinca - get default SYSTIM time increment attributes
3494 * @adapter: board private structure
3495 * @timinca: pointer to returned time increment attributes
3496 *
3497 * Get attributes for incrementing the System Time Register SYSTIML/H at
3498 * the default base frequency, and set the cyclecounter shift value.
3499 **/
3500s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)
3501{
3502	struct e1000_hw *hw = &adapter->hw;
3503	u32 incvalue, incperiod, shift;
3504
3505	/* Make sure clock is enabled on I217/I218/I219  before checking
3506	 * the frequency
3507	 */
3508	if ((hw->mac.type >= e1000_pch_lpt) &&
3509	    !(er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_ENABLED) &&
3510	    !(er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_ENABLED)) {
3511		u32 fextnvm7 = er32(FEXTNVM7);
3512
3513		if (!(fextnvm7 & BIT(0))) {
3514			ew32(FEXTNVM7, fextnvm7 | BIT(0));
3515			e1e_flush();
3516		}
3517	}
3518
3519	switch (hw->mac.type) {
3520	case e1000_pch2lan:
3521		/* Stable 96MHz frequency */
3522		incperiod = INCPERIOD_96MHZ;
3523		incvalue = INCVALUE_96MHZ;
3524		shift = INCVALUE_SHIFT_96MHZ;
3525		adapter->cc.shift = shift + INCPERIOD_SHIFT_96MHZ;
3526		break;
3527	case e1000_pch_lpt:
3528		if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI) {
3529			/* Stable 96MHz frequency */
3530			incperiod = INCPERIOD_96MHZ;
3531			incvalue = INCVALUE_96MHZ;
3532			shift = INCVALUE_SHIFT_96MHZ;
3533			adapter->cc.shift = shift + INCPERIOD_SHIFT_96MHZ;
3534		} else {
3535			/* Stable 25MHz frequency */
3536			incperiod = INCPERIOD_25MHZ;
3537			incvalue = INCVALUE_25MHZ;
3538			shift = INCVALUE_SHIFT_25MHZ;
3539			adapter->cc.shift = shift;
3540		}
3541		break;
3542	case e1000_pch_spt:
3543		/* Stable 24MHz frequency */
3544		incperiod = INCPERIOD_24MHZ;
3545		incvalue = INCVALUE_24MHZ;
3546		shift = INCVALUE_SHIFT_24MHZ;
3547		adapter->cc.shift = shift;
3548		break;
3549	case e1000_pch_cnp:
3550	case e1000_pch_tgp:
3551	case e1000_pch_adp:
3552	case e1000_pch_mtp:
3553		if (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_SYSCFI) {
3554			/* Stable 24MHz frequency */
3555			incperiod = INCPERIOD_24MHZ;
3556			incvalue = INCVALUE_24MHZ;
3557			shift = INCVALUE_SHIFT_24MHZ;
3558			adapter->cc.shift = shift;
3559		} else {
3560			/* Stable 38400KHz frequency */
3561			incperiod = INCPERIOD_38400KHZ;
3562			incvalue = INCVALUE_38400KHZ;
3563			shift = INCVALUE_SHIFT_38400KHZ;
3564			adapter->cc.shift = shift;
3565		}
3566		break;
3567	case e1000_82574:
3568	case e1000_82583:
3569		/* Stable 25MHz frequency */
3570		incperiod = INCPERIOD_25MHZ;
3571		incvalue = INCVALUE_25MHZ;
3572		shift = INCVALUE_SHIFT_25MHZ;
3573		adapter->cc.shift = shift;
3574		break;
3575	default:
3576		return -EINVAL;
3577	}
3578
3579	*timinca = ((incperiod << E1000_TIMINCA_INCPERIOD_SHIFT) |
3580		    ((incvalue << shift) & E1000_TIMINCA_INCVALUE_MASK));
3581
3582	return 0;
3583}
3584
3585/**
3586 * e1000e_config_hwtstamp - configure the hwtstamp registers and enable/disable
3587 * @adapter: board private structure
3588 * @config: timestamp configuration
3589 *
3590 * Outgoing time stamping can be enabled and disabled. Play nice and
3591 * disable it when requested, although it shouldn't cause any overhead
3592 * when no packet needs it. At most one packet in the queue may be
3593 * marked for time stamping, otherwise it would be impossible to tell
3594 * for sure to which packet the hardware time stamp belongs.
3595 *
3596 * Incoming time stamping has to be configured via the hardware filters.
3597 * Not all combinations are supported, in particular event type has to be
3598 * specified. Matching the kind of event packet is not supported, with the
3599 * exception of "all V2 events regardless of level 2 or 4".
3600 **/
3601static int e1000e_config_hwtstamp(struct e1000_adapter *adapter,
3602				  struct hwtstamp_config *config)
3603{
3604	struct e1000_hw *hw = &adapter->hw;
3605	u32 tsync_tx_ctl = E1000_TSYNCTXCTL_ENABLED;
3606	u32 tsync_rx_ctl = E1000_TSYNCRXCTL_ENABLED;
3607	u32 rxmtrl = 0;
3608	u16 rxudp = 0;
3609	bool is_l4 = false;
3610	bool is_l2 = false;
3611	u32 regval;
3612
3613	if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP))
3614		return -EINVAL;
3615
3616	/* flags reserved for future extensions - must be zero */
3617	if (config->flags)
3618		return -EINVAL;
3619
3620	switch (config->tx_type) {
3621	case HWTSTAMP_TX_OFF:
3622		tsync_tx_ctl = 0;
3623		break;
3624	case HWTSTAMP_TX_ON:
3625		break;
3626	default:
3627		return -ERANGE;
3628	}
3629
3630	switch (config->rx_filter) {
3631	case HWTSTAMP_FILTER_NONE:
3632		tsync_rx_ctl = 0;
3633		break;
3634	case HWTSTAMP_FILTER_PTP_V1_L4_SYNC:
3635		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L4_V1;
3636		rxmtrl = E1000_RXMTRL_PTP_V1_SYNC_MESSAGE;
3637		is_l4 = true;
3638		break;
3639	case HWTSTAMP_FILTER_PTP_V1_L4_DELAY_REQ:
3640		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L4_V1;
3641		rxmtrl = E1000_RXMTRL_PTP_V1_DELAY_REQ_MESSAGE;
3642		is_l4 = true;
3643		break;
3644	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
3645		/* Also time stamps V2 L2 Path Delay Request/Response */
3646		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_V2;
3647		rxmtrl = E1000_RXMTRL_PTP_V2_SYNC_MESSAGE;
3648		is_l2 = true;
3649		break;
3650	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
3651		/* Also time stamps V2 L2 Path Delay Request/Response. */
3652		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_V2;
3653		rxmtrl = E1000_RXMTRL_PTP_V2_DELAY_REQ_MESSAGE;
3654		is_l2 = true;
3655		break;
3656	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
3657		/* Hardware cannot filter just V2 L4 Sync messages */
3658		fallthrough;
3659	case HWTSTAMP_FILTER_PTP_V2_SYNC:
3660		/* Also time stamps V2 Path Delay Request/Response. */
3661		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_L4_V2;
3662		rxmtrl = E1000_RXMTRL_PTP_V2_SYNC_MESSAGE;
3663		is_l2 = true;
3664		is_l4 = true;
3665		break;
3666	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
3667		/* Hardware cannot filter just V2 L4 Delay Request messages */
3668		fallthrough;
3669	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
3670		/* Also time stamps V2 Path Delay Request/Response. */
3671		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_L2_L4_V2;
3672		rxmtrl = E1000_RXMTRL_PTP_V2_DELAY_REQ_MESSAGE;
3673		is_l2 = true;
3674		is_l4 = true;
3675		break;
3676	case HWTSTAMP_FILTER_PTP_V2_L4_EVENT:
3677	case HWTSTAMP_FILTER_PTP_V2_L2_EVENT:
3678		/* Hardware cannot filter just V2 L4 or L2 Event messages */
3679		fallthrough;
3680	case HWTSTAMP_FILTER_PTP_V2_EVENT:
3681		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_EVENT_V2;
3682		config->rx_filter = HWTSTAMP_FILTER_PTP_V2_EVENT;
3683		is_l2 = true;
3684		is_l4 = true;
3685		break;
3686	case HWTSTAMP_FILTER_PTP_V1_L4_EVENT:
3687		/* For V1, the hardware can only filter Sync messages or
3688		 * Delay Request messages but not both so fall-through to
3689		 * time stamp all packets.
3690		 */
3691		fallthrough;
3692	case HWTSTAMP_FILTER_NTP_ALL:
3693	case HWTSTAMP_FILTER_ALL:
3694		is_l2 = true;
3695		is_l4 = true;
3696		tsync_rx_ctl |= E1000_TSYNCRXCTL_TYPE_ALL;
3697		config->rx_filter = HWTSTAMP_FILTER_ALL;
3698		break;
3699	default:
3700		return -ERANGE;
3701	}
3702
3703	adapter->hwtstamp_config = *config;
3704
3705	/* enable/disable Tx h/w time stamping */
3706	regval = er32(TSYNCTXCTL);
3707	regval &= ~E1000_TSYNCTXCTL_ENABLED;
3708	regval |= tsync_tx_ctl;
3709	ew32(TSYNCTXCTL, regval);
3710	if ((er32(TSYNCTXCTL) & E1000_TSYNCTXCTL_ENABLED) !=
3711	    (regval & E1000_TSYNCTXCTL_ENABLED)) {
3712		e_err("Timesync Tx Control register not set as expected\n");
3713		return -EAGAIN;
3714	}
3715
3716	/* enable/disable Rx h/w time stamping */
3717	regval = er32(TSYNCRXCTL);
3718	regval &= ~(E1000_TSYNCRXCTL_ENABLED | E1000_TSYNCRXCTL_TYPE_MASK);
3719	regval |= tsync_rx_ctl;
3720	ew32(TSYNCRXCTL, regval);
3721	if ((er32(TSYNCRXCTL) & (E1000_TSYNCRXCTL_ENABLED |
3722				 E1000_TSYNCRXCTL_TYPE_MASK)) !=
3723	    (regval & (E1000_TSYNCRXCTL_ENABLED |
3724		       E1000_TSYNCRXCTL_TYPE_MASK))) {
3725		e_err("Timesync Rx Control register not set as expected\n");
3726		return -EAGAIN;
3727	}
3728
3729	/* L2: define ethertype filter for time stamped packets */
3730	if (is_l2)
3731		rxmtrl |= ETH_P_1588;
3732
3733	/* define which PTP packets get time stamped */
3734	ew32(RXMTRL, rxmtrl);
3735
3736	/* Filter by destination port */
3737	if (is_l4) {
3738		rxudp = PTP_EV_PORT;
3739		cpu_to_be16s(&rxudp);
3740	}
3741	ew32(RXUDP, rxudp);
3742
3743	e1e_flush();
3744
3745	/* Clear TSYNCRXCTL_VALID & TSYNCTXCTL_VALID bit */
3746	er32(RXSTMPH);
3747	er32(TXSTMPH);
3748
3749	return 0;
3750}
3751
3752/**
3753 * e1000_configure - configure the hardware for Rx and Tx
3754 * @adapter: private board structure
3755 **/
3756static void e1000_configure(struct e1000_adapter *adapter)
3757{
3758	struct e1000_ring *rx_ring = adapter->rx_ring;
3759
3760	e1000e_set_rx_mode(adapter->netdev);
3761
3762	e1000_restore_vlan(adapter);
3763	e1000_init_manageability_pt(adapter);
3764
3765	e1000_configure_tx(adapter);
3766
3767	if (adapter->netdev->features & NETIF_F_RXHASH)
3768		e1000e_setup_rss_hash(adapter);
3769	e1000_setup_rctl(adapter);
3770	e1000_configure_rx(adapter);
3771	adapter->alloc_rx_buf(rx_ring, e1000_desc_unused(rx_ring), GFP_KERNEL);
3772}
3773
3774/**
3775 * e1000e_power_up_phy - restore link in case the phy was powered down
3776 * @adapter: address of board private structure
3777 *
3778 * The phy may be powered down to save power and turn off link when the
3779 * driver is unloaded and wake on lan is not enabled (among others)
3780 * *** this routine MUST be followed by a call to e1000e_reset ***
3781 **/
3782void e1000e_power_up_phy(struct e1000_adapter *adapter)
3783{
3784	if (adapter->hw.phy.ops.power_up)
3785		adapter->hw.phy.ops.power_up(&adapter->hw);
3786
3787	adapter->hw.mac.ops.setup_link(&adapter->hw);
3788}
3789
3790/**
3791 * e1000_power_down_phy - Power down the PHY
3792 * @adapter: board private structure
3793 *
3794 * Power down the PHY so no link is implied when interface is down.
3795 * The PHY cannot be powered down if management or WoL is active.
3796 */
3797static void e1000_power_down_phy(struct e1000_adapter *adapter)
3798{
3799	if (adapter->hw.phy.ops.power_down)
3800		adapter->hw.phy.ops.power_down(&adapter->hw);
3801}
3802
3803/**
3804 * e1000_flush_tx_ring - remove all descriptors from the tx_ring
3805 * @adapter: board private structure
3806 *
3807 * We want to clear all pending descriptors from the TX ring.
3808 * zeroing happens when the HW reads the regs. We  assign the ring itself as
3809 * the data of the next descriptor. We don't care about the data we are about
3810 * to reset the HW.
3811 */
3812static void e1000_flush_tx_ring(struct e1000_adapter *adapter)
3813{
3814	struct e1000_hw *hw = &adapter->hw;
3815	struct e1000_ring *tx_ring = adapter->tx_ring;
3816	struct e1000_tx_desc *tx_desc = NULL;
3817	u32 tdt, tctl, txd_lower = E1000_TXD_CMD_IFCS;
3818	u16 size = 512;
3819
3820	tctl = er32(TCTL);
3821	ew32(TCTL, tctl | E1000_TCTL_EN);
3822	tdt = er32(TDT(0));
3823	BUG_ON(tdt != tx_ring->next_to_use);
3824	tx_desc =  E1000_TX_DESC(*tx_ring, tx_ring->next_to_use);
3825	tx_desc->buffer_addr = cpu_to_le64(tx_ring->dma);
3826
3827	tx_desc->lower.data = cpu_to_le32(txd_lower | size);
3828	tx_desc->upper.data = 0;
3829	/* flush descriptors to memory before notifying the HW */
3830	wmb();
3831	tx_ring->next_to_use++;
3832	if (tx_ring->next_to_use == tx_ring->count)
3833		tx_ring->next_to_use = 0;
3834	ew32(TDT(0), tx_ring->next_to_use);
3835	usleep_range(200, 250);
3836}
3837
3838/**
3839 * e1000_flush_rx_ring - remove all descriptors from the rx_ring
3840 * @adapter: board private structure
3841 *
3842 * Mark all descriptors in the RX ring as consumed and disable the rx ring
3843 */
3844static void e1000_flush_rx_ring(struct e1000_adapter *adapter)
3845{
3846	u32 rctl, rxdctl;
3847	struct e1000_hw *hw = &adapter->hw;
3848
3849	rctl = er32(RCTL);
3850	ew32(RCTL, rctl & ~E1000_RCTL_EN);
3851	e1e_flush();
3852	usleep_range(100, 150);
3853
3854	rxdctl = er32(RXDCTL(0));
3855	/* zero the lower 14 bits (prefetch and host thresholds) */
3856	rxdctl &= 0xffffc000;
3857
3858	/* update thresholds: prefetch threshold to 31, host threshold to 1
3859	 * and make sure the granularity is "descriptors" and not "cache lines"
3860	 */
3861	rxdctl |= (0x1F | BIT(8) | E1000_RXDCTL_THRESH_UNIT_DESC);
3862
3863	ew32(RXDCTL(0), rxdctl);
3864	/* momentarily enable the RX ring for the changes to take effect */
3865	ew32(RCTL, rctl | E1000_RCTL_EN);
3866	e1e_flush();
3867	usleep_range(100, 150);
3868	ew32(RCTL, rctl & ~E1000_RCTL_EN);
3869}
3870
3871/**
3872 * e1000_flush_desc_rings - remove all descriptors from the descriptor rings
3873 * @adapter: board private structure
3874 *
3875 * In i219, the descriptor rings must be emptied before resetting the HW
3876 * or before changing the device state to D3 during runtime (runtime PM).
3877 *
3878 * Failure to do this will cause the HW to enter a unit hang state which can
3879 * only be released by PCI reset on the device
3880 *
3881 */
3882
3883static void e1000_flush_desc_rings(struct e1000_adapter *adapter)
3884{
3885	u16 hang_state;
3886	u32 fext_nvm11, tdlen;
3887	struct e1000_hw *hw = &adapter->hw;
3888
3889	/* First, disable MULR fix in FEXTNVM11 */
3890	fext_nvm11 = er32(FEXTNVM11);
3891	fext_nvm11 |= E1000_FEXTNVM11_DISABLE_MULR_FIX;
3892	ew32(FEXTNVM11, fext_nvm11);
3893	/* do nothing if we're not in faulty state, or if the queue is empty */
3894	tdlen = er32(TDLEN(0));
3895	pci_read_config_word(adapter->pdev, PCICFG_DESC_RING_STATUS,
3896			     &hang_state);
3897	if (!(hang_state & FLUSH_DESC_REQUIRED) || !tdlen)
3898		return;
3899	e1000_flush_tx_ring(adapter);
3900	/* recheck, maybe the fault is caused by the rx ring */
3901	pci_read_config_word(adapter->pdev, PCICFG_DESC_RING_STATUS,
3902			     &hang_state);
3903	if (hang_state & FLUSH_DESC_REQUIRED)
3904		e1000_flush_rx_ring(adapter);
3905}
3906
3907/**
3908 * e1000e_systim_reset - reset the timesync registers after a hardware reset
3909 * @adapter: board private structure
3910 *
3911 * When the MAC is reset, all hardware bits for timesync will be reset to the
3912 * default values. This function will restore the settings last in place.
3913 * Since the clock SYSTIME registers are reset, we will simply restore the
3914 * cyclecounter to the kernel real clock time.
3915 **/
3916static void e1000e_systim_reset(struct e1000_adapter *adapter)
3917{
3918	struct ptp_clock_info *info = &adapter->ptp_clock_info;
3919	struct e1000_hw *hw = &adapter->hw;
3920	unsigned long flags;
3921	u32 timinca;
3922	s32 ret_val;
3923
3924	if (!(adapter->flags & FLAG_HAS_HW_TIMESTAMP))
3925		return;
3926
3927	if (info->adjfreq) {
3928		/* restore the previous ptp frequency delta */
3929		ret_val = info->adjfreq(info, adapter->ptp_delta);
3930	} else {
3931		/* set the default base frequency if no adjustment possible */
3932		ret_val = e1000e_get_base_timinca(adapter, &timinca);
3933		if (!ret_val)
3934			ew32(TIMINCA, timinca);
3935	}
3936
3937	if (ret_val) {
3938		dev_warn(&adapter->pdev->dev,
3939			 "Failed to restore TIMINCA clock rate delta: %d\n",
3940			 ret_val);
3941		return;
3942	}
3943
3944	/* reset the systim ns time counter */
3945	spin_lock_irqsave(&adapter->systim_lock, flags);
3946	timecounter_init(&adapter->tc, &adapter->cc,
3947			 ktime_to_ns(ktime_get_real()));
3948	spin_unlock_irqrestore(&adapter->systim_lock, flags);
3949
3950	/* restore the previous hwtstamp configuration settings */
3951	e1000e_config_hwtstamp(adapter, &adapter->hwtstamp_config);
3952}
3953
3954/**
3955 * e1000e_reset - bring the hardware into a known good state
3956 * @adapter: board private structure
3957 *
3958 * This function boots the hardware and enables some settings that
3959 * require a configuration cycle of the hardware - those cannot be
3960 * set/changed during runtime. After reset the device needs to be
3961 * properly configured for Rx, Tx etc.
3962 */
3963void e1000e_reset(struct e1000_adapter *adapter)
3964{
3965	struct e1000_mac_info *mac = &adapter->hw.mac;
3966	struct e1000_fc_info *fc = &adapter->hw.fc;
3967	struct e1000_hw *hw = &adapter->hw;
3968	u32 tx_space, min_tx_space, min_rx_space;
3969	u32 pba = adapter->pba;
3970	u16 hwm;
3971
3972	/* reset Packet Buffer Allocation to default */
3973	ew32(PBA, pba);
3974
3975	if (adapter->max_frame_size > (VLAN_ETH_FRAME_LEN + ETH_FCS_LEN)) {
3976		/* To maintain wire speed transmits, the Tx FIFO should be
3977		 * large enough to accommodate two full transmit packets,
3978		 * rounded up to the next 1KB and expressed in KB.  Likewise,
3979		 * the Rx FIFO should be large enough to accommodate at least
3980		 * one full receive packet and is similarly rounded up and
3981		 * expressed in KB.
3982		 */
3983		pba = er32(PBA);
3984		/* upper 16 bits has Tx packet buffer allocation size in KB */
3985		tx_space = pba >> 16;
3986		/* lower 16 bits has Rx packet buffer allocation size in KB */
3987		pba &= 0xffff;
3988		/* the Tx fifo also stores 16 bytes of information about the Tx
3989		 * but don't include ethernet FCS because hardware appends it
3990		 */
3991		min_tx_space = (adapter->max_frame_size +
3992				sizeof(struct e1000_tx_desc) - ETH_FCS_LEN) * 2;
3993		min_tx_space = ALIGN(min_tx_space, 1024);
3994		min_tx_space >>= 10;
3995		/* software strips receive CRC, so leave room for it */
3996		min_rx_space = adapter->max_frame_size;
3997		min_rx_space = ALIGN(min_rx_space, 1024);
3998		min_rx_space >>= 10;
3999
4000		/* If current Tx allocation is less than the min Tx FIFO size,
4001		 * and the min Tx FIFO size is less than the current Rx FIFO
4002		 * allocation, take space away from current Rx allocation
4003		 */
4004		if ((tx_space < min_tx_space) &&
4005		    ((min_tx_space - tx_space) < pba)) {
4006			pba -= min_tx_space - tx_space;
4007
4008			/* if short on Rx space, Rx wins and must trump Tx
4009			 * adjustment
4010			 */
4011			if (pba < min_rx_space)
4012				pba = min_rx_space;
4013		}
4014
4015		ew32(PBA, pba);
4016	}
4017
4018	/* flow control settings
4019	 *
4020	 * The high water mark must be low enough to fit one full frame
4021	 * (or the size used for early receive) above it in the Rx FIFO.
4022	 * Set it to the lower of:
4023	 * - 90% of the Rx FIFO size, and
4024	 * - the full Rx FIFO size minus one full frame
4025	 */
4026	if (adapter->flags & FLAG_DISABLE_FC_PAUSE_TIME)
4027		fc->pause_time = 0xFFFF;
4028	else
4029		fc->pause_time = E1000_FC_PAUSE_TIME;
4030	fc->send_xon = true;
4031	fc->current_mode = fc->requested_mode;
4032
4033	switch (hw->mac.type) {
4034	case e1000_ich9lan:
4035	case e1000_ich10lan:
4036		if (adapter->netdev->mtu > ETH_DATA_LEN) {
4037			pba = 14;
4038			ew32(PBA, pba);
4039			fc->high_water = 0x2800;
4040			fc->low_water = fc->high_water - 8;
4041			break;
4042		}
4043		fallthrough;
4044	default:
4045		hwm = min(((pba << 10) * 9 / 10),
4046			  ((pba << 10) - adapter->max_frame_size));
4047
4048		fc->high_water = hwm & E1000_FCRTH_RTH;	/* 8-byte granularity */
4049		fc->low_water = fc->high_water - 8;
4050		break;
4051	case e1000_pchlan:
4052		/* Workaround PCH LOM adapter hangs with certain network
4053		 * loads.  If hangs persist, try disabling Tx flow control.
4054		 */
4055		if (adapter->netdev->mtu > ETH_DATA_LEN) {
4056			fc->high_water = 0x3500;
4057			fc->low_water = 0x1500;
4058		} else {
4059			fc->high_water = 0x5000;
4060			fc->low_water = 0x3000;
4061		}
4062		fc->refresh_time = 0x1000;
4063		break;
4064	case e1000_pch2lan:
4065	case e1000_pch_lpt:
4066	case e1000_pch_spt:
4067	case e1000_pch_cnp:
4068	case e1000_pch_tgp:
4069	case e1000_pch_adp:
4070	case e1000_pch_mtp:
4071		fc->refresh_time = 0xFFFF;
4072		fc->pause_time = 0xFFFF;
4073
4074		if (adapter->netdev->mtu <= ETH_DATA_LEN) {
4075			fc->high_water = 0x05C20;
4076			fc->low_water = 0x05048;
4077			break;
4078		}
4079
4080		pba = 14;
4081		ew32(PBA, pba);
4082		fc->high_water = ((pba << 10) * 9 / 10) & E1000_FCRTH_RTH;
4083		fc->low_water = ((pba << 10) * 8 / 10) & E1000_FCRTL_RTL;
4084		break;
4085	}
4086
4087	/* Alignment of Tx data is on an arbitrary byte boundary with the
4088	 * maximum size per Tx descriptor limited only to the transmit
4089	 * allocation of the packet buffer minus 96 bytes with an upper
4090	 * limit of 24KB due to receive synchronization limitations.
4091	 */
4092	adapter->tx_fifo_limit = min_t(u32, ((er32(PBA) >> 16) << 10) - 96,
4093				       24 << 10);
4094
4095	/* Disable Adaptive Interrupt Moderation if 2 full packets cannot
4096	 * fit in receive buffer.
4097	 */
4098	if (adapter->itr_setting & 0x3) {
4099		if ((adapter->max_frame_size * 2) > (pba << 10)) {
4100			if (!(adapter->flags2 & FLAG2_DISABLE_AIM)) {
4101				dev_info(&adapter->pdev->dev,
4102					 "Interrupt Throttle Rate off\n");
4103				adapter->flags2 |= FLAG2_DISABLE_AIM;
4104				e1000e_write_itr(adapter, 0);
4105			}
4106		} else if (adapter->flags2 & FLAG2_DISABLE_AIM) {
4107			dev_info(&adapter->pdev->dev,
4108				 "Interrupt Throttle Rate on\n");
4109			adapter->flags2 &= ~FLAG2_DISABLE_AIM;
4110			adapter->itr = 20000;
4111			e1000e_write_itr(adapter, adapter->itr);
4112		}
4113	}
4114
4115	if (hw->mac.type >= e1000_pch_spt)
4116		e1000_flush_desc_rings(adapter);
4117	/* Allow time for pending master requests to run */
4118	mac->ops.reset_hw(hw);
4119
4120	/* For parts with AMT enabled, let the firmware know
4121	 * that the network interface is in control
4122	 */
4123	if (adapter->flags & FLAG_HAS_AMT)
4124		e1000e_get_hw_control(adapter);
4125
4126	ew32(WUC, 0);
4127
4128	if (mac->ops.init_hw(hw))
4129		e_err("Hardware Error\n");
4130
4131	e1000_update_mng_vlan(adapter);
4132
4133	/* Enable h/w to recognize an 802.1Q VLAN Ethernet packet */
4134	ew32(VET, ETH_P_8021Q);
4135
4136	e1000e_reset_adaptive(hw);
4137
4138	/* restore systim and hwtstamp settings */
4139	e1000e_systim_reset(adapter);
4140
4141	/* Set EEE advertisement as appropriate */
4142	if (adapter->flags2 & FLAG2_HAS_EEE) {
4143		s32 ret_val;
4144		u16 adv_addr;
4145
4146		switch (hw->phy.type) {
4147		case e1000_phy_82579:
4148			adv_addr = I82579_EEE_ADVERTISEMENT;
4149			break;
4150		case e1000_phy_i217:
4151			adv_addr = I217_EEE_ADVERTISEMENT;
4152			break;
4153		default:
4154			dev_err(&adapter->pdev->dev,
4155				"Invalid PHY type setting EEE advertisement\n");
4156			return;
4157		}
4158
4159		ret_val = hw->phy.ops.acquire(hw);
4160		if (ret_val) {
4161			dev_err(&adapter->pdev->dev,
4162				"EEE advertisement - unable to acquire PHY\n");
4163			return;
4164		}
4165
4166		e1000_write_emi_reg_locked(hw, adv_addr,
4167					   hw->dev_spec.ich8lan.eee_disable ?
4168					   0 : adapter->eee_advert);
4169
4170		hw->phy.ops.release(hw);
4171	}
4172
4173	if (!netif_running(adapter->netdev) &&
4174	    !test_bit(__E1000_TESTING, &adapter->state))
4175		e1000_power_down_phy(adapter);
4176
4177	e1000_get_phy_info(hw);
4178
4179	if ((adapter->flags & FLAG_HAS_SMART_POWER_DOWN) &&
4180	    !(adapter->flags & FLAG_SMART_POWER_DOWN)) {
4181		u16 phy_data = 0;
4182		/* speed up time to link by disabling smart power down, ignore
4183		 * the return value of this function because there is nothing
4184		 * different we would do if it failed
4185		 */
4186		e1e_rphy(hw, IGP02E1000_PHY_POWER_MGMT, &phy_data);
4187		phy_data &= ~IGP02E1000_PM_SPD;
4188		e1e_wphy(hw, IGP02E1000_PHY_POWER_MGMT, phy_data);
4189	}
4190	if (hw->mac.type >= e1000_pch_spt && adapter->int_mode == 0) {
4191		u32 reg;
4192
4193		/* Fextnvm7 @ 0xe4[2] = 1 */
4194		reg = er32(FEXTNVM7);
4195		reg |= E1000_FEXTNVM7_SIDE_CLK_UNGATE;
4196		ew32(FEXTNVM7, reg);
4197		/* Fextnvm9 @ 0x5bb4[13:12] = 11 */
4198		reg = er32(FEXTNVM9);
4199		reg |= E1000_FEXTNVM9_IOSFSB_CLKGATE_DIS |
4200		       E1000_FEXTNVM9_IOSFSB_CLKREQ_DIS;
4201		ew32(FEXTNVM9, reg);
4202	}
4203
4204}
4205
4206/**
4207 * e1000e_trigger_lsc - trigger an LSC interrupt
4208 * @adapter:
4209 *
4210 * Fire a link status change interrupt to start the watchdog.
4211 **/
4212static void e1000e_trigger_lsc(struct e1000_adapter *adapter)
4213{
4214	struct e1000_hw *hw = &adapter->hw;
4215
4216	if (adapter->msix_entries)
4217		ew32(ICS, E1000_ICS_LSC | E1000_ICS_OTHER);
4218	else
4219		ew32(ICS, E1000_ICS_LSC);
4220}
4221
4222void e1000e_up(struct e1000_adapter *adapter)
4223{
4224	/* hardware has been reset, we need to reload some things */
4225	e1000_configure(adapter);
4226
4227	clear_bit(__E1000_DOWN, &adapter->state);
4228
4229	if (adapter->msix_entries)
4230		e1000_configure_msix(adapter);
4231	e1000_irq_enable(adapter);
4232
4233	/* Tx queue started by watchdog timer when link is up */
4234
4235	e1000e_trigger_lsc(adapter);
4236}
4237
4238static void e1000e_flush_descriptors(struct e1000_adapter *adapter)
4239{
4240	struct e1000_hw *hw = &adapter->hw;
4241
4242	if (!(adapter->flags2 & FLAG2_DMA_BURST))
4243		return;
4244
4245	/* flush pending descriptor writebacks to memory */
4246	ew32(TIDV, adapter->tx_int_delay | E1000_TIDV_FPD);
4247	ew32(RDTR, adapter->rx_int_delay | E1000_RDTR_FPD);
4248
4249	/* execute the writes immediately */
4250	e1e_flush();
4251
4252	/* due to rare timing issues, write to TIDV/RDTR again to ensure the
4253	 * write is successful
4254	 */
4255	ew32(TIDV, adapter->tx_int_delay | E1000_TIDV_FPD);
4256	ew32(RDTR, adapter->rx_int_delay | E1000_RDTR_FPD);
4257
4258	/* execute the writes immediately */
4259	e1e_flush();
4260}
4261
4262static void e1000e_update_stats(struct e1000_adapter *adapter);
4263
4264/**
4265 * e1000e_down - quiesce the device and optionally reset the hardware
4266 * @adapter: board private structure
4267 * @reset: boolean flag to reset the hardware or not
4268 */
4269void e1000e_down(struct e1000_adapter *adapter, bool reset)
4270{
4271	struct net_device *netdev = adapter->netdev;
4272	struct e1000_hw *hw = &adapter->hw;
4273	u32 tctl, rctl;
4274
4275	/* signal that we're down so the interrupt handler does not
4276	 * reschedule our watchdog timer
4277	 */
4278	set_bit(__E1000_DOWN, &adapter->state);
4279
4280	netif_carrier_off(netdev);
4281
4282	/* disable receives in the hardware */
4283	rctl = er32(RCTL);
4284	if (!(adapter->flags2 & FLAG2_NO_DISABLE_RX))
4285		ew32(RCTL, rctl & ~E1000_RCTL_EN);
4286	/* flush and sleep below */
4287
4288	netif_stop_queue(netdev);
4289
4290	/* disable transmits in the hardware */
4291	tctl = er32(TCTL);
4292	tctl &= ~E1000_TCTL_EN;
4293	ew32(TCTL, tctl);
4294
4295	/* flush both disables and wait for them to finish */
4296	e1e_flush();
4297	usleep_range(10000, 11000);
4298
4299	e1000_irq_disable(adapter);
4300
4301	napi_synchronize(&adapter->napi);
4302
4303	del_timer_sync(&adapter->watchdog_timer);
4304	del_timer_sync(&adapter->phy_info_timer);
4305
4306	spin_lock(&adapter->stats64_lock);
4307	e1000e_update_stats(adapter);
4308	spin_unlock(&adapter->stats64_lock);
4309
4310	e1000e_flush_descriptors(adapter);
4311
4312	adapter->link_speed = 0;
4313	adapter->link_duplex = 0;
4314
4315	/* Disable Si errata workaround on PCHx for jumbo frame flow */
4316	if ((hw->mac.type >= e1000_pch2lan) &&
4317	    (adapter->netdev->mtu > ETH_DATA_LEN) &&
4318	    e1000_lv_jumbo_workaround_ich8lan(hw, false))
4319		e_dbg("failed to disable jumbo frame workaround mode\n");
4320
4321	if (!pci_channel_offline(adapter->pdev)) {
4322		if (reset)
4323			e1000e_reset(adapter);
4324		else if (hw->mac.type >= e1000_pch_spt)
4325			e1000_flush_desc_rings(adapter);
4326	}
4327	e1000_clean_tx_ring(adapter->tx_ring);
4328	e1000_clean_rx_ring(adapter->rx_ring);
4329}
4330
4331void e1000e_reinit_locked(struct e1000_adapter *adapter)
4332{
4333	might_sleep();
4334	while (test_and_set_bit(__E1000_RESETTING, &adapter->state))
4335		usleep_range(1000, 1100);
4336	e1000e_down(adapter, true);
4337	e1000e_up(adapter);
4338	clear_bit(__E1000_RESETTING, &adapter->state);
4339}
4340
4341/**
4342 * e1000e_sanitize_systim - sanitize raw cycle counter reads
4343 * @hw: pointer to the HW structure
4344 * @systim: PHC time value read, sanitized and returned
4345 * @sts: structure to hold system time before and after reading SYSTIML,
4346 * may be NULL
4347 *
4348 * Errata for 82574/82583 possible bad bits read from SYSTIMH/L:
4349 * check to see that the time is incrementing at a reasonable
4350 * rate and is a multiple of incvalue.
4351 **/
4352static u64 e1000e_sanitize_systim(struct e1000_hw *hw, u64 systim,
4353				  struct ptp_system_timestamp *sts)
4354{
4355	u64 time_delta, rem, temp;
4356	u64 systim_next;
4357	u32 incvalue;
4358	int i;
4359
4360	incvalue = er32(TIMINCA) & E1000_TIMINCA_INCVALUE_MASK;
4361	for (i = 0; i < E1000_MAX_82574_SYSTIM_REREADS; i++) {
4362		/* latch SYSTIMH on read of SYSTIML */
4363		ptp_read_system_prets(sts);
4364		systim_next = (u64)er32(SYSTIML);
4365		ptp_read_system_postts(sts);
4366		systim_next |= (u64)er32(SYSTIMH) << 32;
4367
4368		time_delta = systim_next - systim;
4369		temp = time_delta;
4370		/* VMWare users have seen incvalue of zero, don't div / 0 */
4371		rem = incvalue ? do_div(temp, incvalue) : (time_delta != 0);
4372
4373		systim = systim_next;
4374
4375		if ((time_delta < E1000_82574_SYSTIM_EPSILON) && (rem == 0))
4376			break;
4377	}
4378
4379	return systim;
4380}
4381
4382/**
4383 * e1000e_read_systim - read SYSTIM register
4384 * @adapter: board private structure
4385 * @sts: structure which will contain system time before and after reading
4386 * SYSTIML, may be NULL
4387 **/
4388u64 e1000e_read_systim(struct e1000_adapter *adapter,
4389		       struct ptp_system_timestamp *sts)
4390{
4391	struct e1000_hw *hw = &adapter->hw;
4392	u32 systimel, systimel_2, systimeh;
4393	u64 systim;
4394	/* SYSTIMH latching upon SYSTIML read does not work well.
4395	 * This means that if SYSTIML overflows after we read it but before
4396	 * we read SYSTIMH, the value of SYSTIMH has been incremented and we
4397	 * will experience a huge non linear increment in the systime value
4398	 * to fix that we test for overflow and if true, we re-read systime.
4399	 */
4400	ptp_read_system_prets(sts);
4401	systimel = er32(SYSTIML);
4402	ptp_read_system_postts(sts);
4403	systimeh = er32(SYSTIMH);
4404	/* Is systimel is so large that overflow is possible? */
4405	if (systimel >= (u32)0xffffffff - E1000_TIMINCA_INCVALUE_MASK) {
4406		ptp_read_system_prets(sts);
4407		systimel_2 = er32(SYSTIML);
4408		ptp_read_system_postts(sts);
4409		if (systimel > systimel_2) {
4410			/* There was an overflow, read again SYSTIMH, and use
4411			 * systimel_2
4412			 */
4413			systimeh = er32(SYSTIMH);
4414			systimel = systimel_2;
4415		}
4416	}
4417	systim = (u64)systimel;
4418	systim |= (u64)systimeh << 32;
4419
4420	if (adapter->flags2 & FLAG2_CHECK_SYSTIM_OVERFLOW)
4421		systim = e1000e_sanitize_systim(hw, systim, sts);
4422
4423	return systim;
4424}
4425
4426/**
4427 * e1000e_cyclecounter_read - read raw cycle counter (used by time counter)
4428 * @cc: cyclecounter structure
4429 **/
4430static u64 e1000e_cyclecounter_read(const struct cyclecounter *cc)
4431{
4432	struct e1000_adapter *adapter = container_of(cc, struct e1000_adapter,
4433						     cc);
4434
4435	return e1000e_read_systim(adapter, NULL);
4436}
4437
4438/**
4439 * e1000_sw_init - Initialize general software structures (struct e1000_adapter)
4440 * @adapter: board private structure to initialize
4441 *
4442 * e1000_sw_init initializes the Adapter private data structure.
4443 * Fields are initialized based on PCI device information and
4444 * OS network device settings (MTU size).
4445 **/
4446static int e1000_sw_init(struct e1000_adapter *adapter)
4447{
4448	struct net_device *netdev = adapter->netdev;
4449
4450	adapter->rx_buffer_len = VLAN_ETH_FRAME_LEN + ETH_FCS_LEN;
4451	adapter->rx_ps_bsize0 = 128;
4452	adapter->max_frame_size = netdev->mtu + VLAN_ETH_HLEN + ETH_FCS_LEN;
4453	adapter->min_frame_size = ETH_ZLEN + ETH_FCS_LEN;
4454	adapter->tx_ring_count = E1000_DEFAULT_TXD;
4455	adapter->rx_ring_count = E1000_DEFAULT_RXD;
4456
4457	spin_lock_init(&adapter->stats64_lock);
4458
4459	e1000e_set_interrupt_capability(adapter);
4460
4461	if (e1000_alloc_queues(adapter))
4462		return -ENOMEM;
4463
4464	/* Setup hardware time stamping cyclecounter */
4465	if (adapter->flags & FLAG_HAS_HW_TIMESTAMP) {
4466		adapter->cc.read = e1000e_cyclecounter_read;
4467		adapter->cc.mask = CYCLECOUNTER_MASK(64);
4468		adapter->cc.mult = 1;
4469		/* cc.shift set in e1000e_get_base_tininca() */
4470
4471		spin_lock_init(&adapter->systim_lock);
4472		INIT_WORK(&adapter->tx_hwtstamp_work, e1000e_tx_hwtstamp_work);
4473	}
4474
4475	/* Explicitly disable IRQ since the NIC can be in any state. */
4476	e1000_irq_disable(adapter);
4477
4478	set_bit(__E1000_DOWN, &adapter->state);
4479	return 0;
4480}
4481
4482/**
4483 * e1000_intr_msi_test - Interrupt Handler
4484 * @irq: interrupt number
4485 * @data: pointer to a network interface device structure
4486 **/
4487static irqreturn_t e1000_intr_msi_test(int __always_unused irq, void *data)
4488{
4489	struct net_device *netdev = data;
4490	struct e1000_adapter *adapter = netdev_priv(netdev);
4491	struct e1000_hw *hw = &adapter->hw;
4492	u32 icr = er32(ICR);
4493
4494	e_dbg("icr is %08X\n", icr);
4495	if (icr & E1000_ICR_RXSEQ) {
4496		adapter->flags &= ~FLAG_MSI_TEST_FAILED;
4497		/* Force memory writes to complete before acknowledging the
4498		 * interrupt is handled.
4499		 */
4500		wmb();
4501	}
4502
4503	return IRQ_HANDLED;
4504}
4505
4506/**
4507 * e1000_test_msi_interrupt - Returns 0 for successful test
4508 * @adapter: board private struct
4509 *
4510 * code flow taken from tg3.c
4511 **/
4512static int e1000_test_msi_interrupt(struct e1000_adapter *adapter)
4513{
4514	struct net_device *netdev = adapter->netdev;
4515	struct e1000_hw *hw = &adapter->hw;
4516	int err;
4517
4518	/* poll_enable hasn't been called yet, so don't need disable */
4519	/* clear any pending events */
4520	er32(ICR);
4521
4522	/* free the real vector and request a test handler */
4523	e1000_free_irq(adapter);
4524	e1000e_reset_interrupt_capability(adapter);
4525
4526	/* Assume that the test fails, if it succeeds then the test
4527	 * MSI irq handler will unset this flag
4528	 */
4529	adapter->flags |= FLAG_MSI_TEST_FAILED;
4530
4531	err = pci_enable_msi(adapter->pdev);
4532	if (err)
4533		goto msi_test_failed;
4534
4535	err = request_irq(adapter->pdev->irq, e1000_intr_msi_test, 0,
4536			  netdev->name, netdev);
4537	if (err) {
4538		pci_disable_msi(adapter->pdev);
4539		goto msi_test_failed;
4540	}
4541
4542	/* Force memory writes to complete before enabling and firing an
4543	 * interrupt.
4544	 */
4545	wmb();
4546
4547	e1000_irq_enable(adapter);
4548
4549	/* fire an unusual interrupt on the test handler */
4550	ew32(ICS, E1000_ICS_RXSEQ);
4551	e1e_flush();
4552	msleep(100);
4553
4554	e1000_irq_disable(adapter);
4555
4556	rmb();			/* read flags after interrupt has been fired */
4557
4558	if (adapter->flags & FLAG_MSI_TEST_FAILED) {
4559		adapter->int_mode = E1000E_INT_MODE_LEGACY;
4560		e_info("MSI interrupt test failed, using legacy interrupt.\n");
4561	} else {
4562		e_dbg("MSI interrupt test succeeded!\n");
4563	}
4564
4565	free_irq(adapter->pdev->irq, netdev);
4566	pci_disable_msi(adapter->pdev);
4567
4568msi_test_failed:
4569	e1000e_set_interrupt_capability(adapter);
4570	return e1000_request_irq(adapter);
4571}
4572
4573/**
4574 * e1000_test_msi - Returns 0 if MSI test succeeds or INTx mode is restored
4575 * @adapter: board private struct
4576 *
4577 * code flow taken from tg3.c, called with e1000 interrupts disabled.
4578 **/
4579static int e1000_test_msi(struct e1000_adapter *adapter)
4580{
4581	int err;
4582	u16 pci_cmd;
4583
4584	if (!(adapter->flags & FLAG_MSI_ENABLED))
4585		return 0;
4586
4587	/* disable SERR in case the MSI write causes a master abort */
4588	pci_read_config_word(adapter->pdev, PCI_COMMAND, &pci_cmd);
4589	if (pci_cmd & PCI_COMMAND_SERR)
4590		pci_write_config_word(adapter->pdev, PCI_COMMAND,
4591				      pci_cmd & ~PCI_COMMAND_SERR);
4592
4593	err = e1000_test_msi_interrupt(adapter);
4594
4595	/* re-enable SERR */
4596	if (pci_cmd & PCI_COMMAND_SERR) {
4597		pci_read_config_word(adapter->pdev, PCI_COMMAND, &pci_cmd);
4598		pci_cmd |= PCI_COMMAND_SERR;
4599		pci_write_config_word(adapter->pdev, PCI_COMMAND, pci_cmd);
4600	}
4601
4602	return err;
4603}
4604
4605/**
4606 * e1000e_open - Called when a network interface is made active
4607 * @netdev: network interface device structure
4608 *
4609 * Returns 0 on success, negative value on failure
4610 *
4611 * The open entry point is called when a network interface is made
4612 * active by the system (IFF_UP).  At this point all resources needed
4613 * for transmit and receive operations are allocated, the interrupt
4614 * handler is registered with the OS, the watchdog timer is started,
4615 * and the stack is notified that the interface is ready.
4616 **/
4617int e1000e_open(struct net_device *netdev)
4618{
4619	struct e1000_adapter *adapter = netdev_priv(netdev);
4620	struct e1000_hw *hw = &adapter->hw;
4621	struct pci_dev *pdev = adapter->pdev;
4622	int err;
4623
4624	/* disallow open during test */
4625	if (test_bit(__E1000_TESTING, &adapter->state))
4626		return -EBUSY;
4627
4628	pm_runtime_get_sync(&pdev->dev);
4629
4630	netif_carrier_off(netdev);
4631	netif_stop_queue(netdev);
4632
4633	/* allocate transmit descriptors */
4634	err = e1000e_setup_tx_resources(adapter->tx_ring);
4635	if (err)
4636		goto err_setup_tx;
4637
4638	/* allocate receive descriptors */
4639	err = e1000e_setup_rx_resources(adapter->rx_ring);
4640	if (err)
4641		goto err_setup_rx;
4642
4643	/* If AMT is enabled, let the firmware know that the network
4644	 * interface is now open and reset the part to a known state.
4645	 */
4646	if (adapter->flags & FLAG_HAS_AMT) {
4647		e1000e_get_hw_control(adapter);
4648		e1000e_reset(adapter);
4649	}
4650
4651	e1000e_power_up_phy(adapter);
4652
4653	adapter->mng_vlan_id = E1000_MNG_VLAN_NONE;
4654	if ((adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN))
4655		e1000_update_mng_vlan(adapter);
4656
4657	/* DMA latency requirement to workaround jumbo issue */
4658	cpu_latency_qos_add_request(&adapter->pm_qos_req, PM_QOS_DEFAULT_VALUE);
4659
4660	/* before we allocate an interrupt, we must be ready to handle it.
4661	 * Setting DEBUG_SHIRQ in the kernel makes it fire an interrupt
4662	 * as soon as we call pci_request_irq, so we have to setup our
4663	 * clean_rx handler before we do so.
4664	 */
4665	e1000_configure(adapter);
4666
4667	err = e1000_request_irq(adapter);
4668	if (err)
4669		goto err_req_irq;
4670
4671	/* Work around PCIe errata with MSI interrupts causing some chipsets to
4672	 * ignore e1000e MSI messages, which means we need to test our MSI
4673	 * interrupt now
4674	 */
4675	if (adapter->int_mode != E1000E_INT_MODE_LEGACY) {
4676		err = e1000_test_msi(adapter);
4677		if (err) {
4678			e_err("Interrupt allocation failed\n");
4679			goto err_req_irq;
4680		}
4681	}
4682
4683	/* From here on the code is the same as e1000e_up() */
4684	clear_bit(__E1000_DOWN, &adapter->state);
4685
4686	napi_enable(&adapter->napi);
4687
4688	e1000_irq_enable(adapter);
4689
4690	adapter->tx_hang_recheck = false;
4691
4692	hw->mac.get_link_status = true;
4693	pm_runtime_put(&pdev->dev);
4694
4695	e1000e_trigger_lsc(adapter);
4696
4697	return 0;
4698
4699err_req_irq:
4700	cpu_latency_qos_remove_request(&adapter->pm_qos_req);
4701	e1000e_release_hw_control(adapter);
4702	e1000_power_down_phy(adapter);
4703	e1000e_free_rx_resources(adapter->rx_ring);
4704err_setup_rx:
4705	e1000e_free_tx_resources(adapter->tx_ring);
4706err_setup_tx:
4707	e1000e_reset(adapter);
4708	pm_runtime_put_sync(&pdev->dev);
4709
4710	return err;
4711}
4712
4713/**
4714 * e1000e_close - Disables a network interface
4715 * @netdev: network interface device structure
4716 *
4717 * Returns 0, this is not allowed to fail
4718 *
4719 * The close entry point is called when an interface is de-activated
4720 * by the OS.  The hardware is still under the drivers control, but
4721 * needs to be disabled.  A global MAC reset is issued to stop the
4722 * hardware, and all transmit and receive resources are freed.
4723 **/
4724int e1000e_close(struct net_device *netdev)
4725{
4726	struct e1000_adapter *adapter = netdev_priv(netdev);
4727	struct pci_dev *pdev = adapter->pdev;
4728	int count = E1000_CHECK_RESET_COUNT;
4729
4730	while (test_bit(__E1000_RESETTING, &adapter->state) && count--)
4731		usleep_range(10000, 11000);
4732
4733	WARN_ON(test_bit(__E1000_RESETTING, &adapter->state));
4734
4735	pm_runtime_get_sync(&pdev->dev);
4736
4737	if (netif_device_present(netdev)) {
4738		e1000e_down(adapter, true);
4739		e1000_free_irq(adapter);
4740
4741		/* Link status message must follow this format */
4742		netdev_info(netdev, "NIC Link is Down\n");
4743	}
4744
4745	napi_disable(&adapter->napi);
4746
4747	e1000e_free_tx_resources(adapter->tx_ring);
4748	e1000e_free_rx_resources(adapter->rx_ring);
4749
4750	/* kill manageability vlan ID if supported, but not if a vlan with
4751	 * the same ID is registered on the host OS (let 8021q kill it)
4752	 */
4753	if (adapter->hw.mng_cookie.status & E1000_MNG_DHCP_COOKIE_STATUS_VLAN)
4754		e1000_vlan_rx_kill_vid(netdev, htons(ETH_P_8021Q),
4755				       adapter->mng_vlan_id);
4756
4757	/* If AMT is enabled, let the firmware know that the network
4758	 * interface is now closed
4759	 */
4760	if ((adapter->flags & FLAG_HAS_AMT) &&
4761	    !test_bit(__E1000_TESTING, &adapter->state))
4762		e1000e_release_hw_control(adapter);
4763
4764	cpu_latency_qos_remove_request(&adapter->pm_qos_req);
4765
4766	pm_runtime_put_sync(&pdev->dev);
4767
4768	return 0;
4769}
4770
4771/**
4772 * e1000_set_mac - Change the Ethernet Address of the NIC
4773 * @netdev: network interface device structure
4774 * @p: pointer to an address structure
4775 *
4776 * Returns 0 on success, negative on failure
4777 **/
4778static int e1000_set_mac(struct net_device *netdev, void *p)
4779{
4780	struct e1000_adapter *adapter = netdev_priv(netdev);
4781	struct e1000_hw *hw = &adapter->hw;
4782	struct sockaddr *addr = p;
4783
4784	if (!is_valid_ether_addr(addr->sa_data))
4785		return -EADDRNOTAVAIL;
4786
4787	memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
4788	memcpy(adapter->hw.mac.addr, addr->sa_data, netdev->addr_len);
4789
4790	hw->mac.ops.rar_set(&adapter->hw, adapter->hw.mac.addr, 0);
4791
4792	if (adapter->flags & FLAG_RESET_OVERWRITES_LAA) {
4793		/* activate the work around */
4794		e1000e_set_laa_state_82571(&adapter->hw, 1);
4795
4796		/* Hold a copy of the LAA in RAR[14] This is done so that
4797		 * between the time RAR[0] gets clobbered  and the time it
4798		 * gets fixed (in e1000_watchdog), the actual LAA is in one
4799		 * of the RARs and no incoming packets directed to this port
4800		 * are dropped. Eventually the LAA will be in RAR[0] and
4801		 * RAR[14]
4802		 */
4803		hw->mac.ops.rar_set(&adapter->hw, adapter->hw.mac.addr,
4804				    adapter->hw.mac.rar_entry_count - 1);
4805	}
4806
4807	return 0;
4808}
4809
4810/**
4811 * e1000e_update_phy_task - work thread to update phy
4812 * @work: pointer to our work struct
4813 *
4814 * this worker thread exists because we must acquire a
4815 * semaphore to read the phy, which we could msleep while
4816 * waiting for it, and we can't msleep in a timer.
4817 **/
4818static void e1000e_update_phy_task(struct work_struct *work)
4819{
4820	struct e1000_adapter *adapter = container_of(work,
4821						     struct e1000_adapter,
4822						     update_phy_task);
4823	struct e1000_hw *hw = &adapter->hw;
4824
4825	if (test_bit(__E1000_DOWN, &adapter->state))
4826		return;
4827
4828	e1000_get_phy_info(hw);
4829
4830	/* Enable EEE on 82579 after link up */
4831	if (hw->phy.type >= e1000_phy_82579)
4832		e1000_set_eee_pchlan(hw);
4833}
4834
4835/**
4836 * e1000_update_phy_info - timre call-back to update PHY info
4837 * @t: pointer to timer_list containing private info adapter
4838 *
4839 * Need to wait a few seconds after link up to get diagnostic information from
4840 * the phy
4841 **/
4842static void e1000_update_phy_info(struct timer_list *t)
4843{
4844	struct e1000_adapter *adapter = from_timer(adapter, t, phy_info_timer);
4845
4846	if (test_bit(__E1000_DOWN, &adapter->state))
4847		return;
4848
4849	schedule_work(&adapter->update_phy_task);
4850}
4851
4852/**
4853 * e1000e_update_phy_stats - Update the PHY statistics counters
4854 * @adapter: board private structure
4855 *
4856 * Read/clear the upper 16-bit PHY registers and read/accumulate lower
4857 **/
4858static void e1000e_update_phy_stats(struct e1000_adapter *adapter)
4859{
4860	struct e1000_hw *hw = &adapter->hw;
4861	s32 ret_val;
4862	u16 phy_data;
4863
4864	ret_val = hw->phy.ops.acquire(hw);
4865	if (ret_val)
4866		return;
4867
4868	/* A page set is expensive so check if already on desired page.
4869	 * If not, set to the page with the PHY status registers.
4870	 */
4871	hw->phy.addr = 1;
4872	ret_val = e1000e_read_phy_reg_mdic(hw, IGP01E1000_PHY_PAGE_SELECT,
4873					   &phy_data);
4874	if (ret_val)
4875		goto release;
4876	if (phy_data != (HV_STATS_PAGE << IGP_PAGE_SHIFT)) {
4877		ret_val = hw->phy.ops.set_page(hw,
4878					       HV_STATS_PAGE << IGP_PAGE_SHIFT);
4879		if (ret_val)
4880			goto release;
4881	}
4882
4883	/* Single Collision Count */
4884	hw->phy.ops.read_reg_page(hw, HV_SCC_UPPER, &phy_data);
4885	ret_val = hw->phy.ops.read_reg_page(hw, HV_SCC_LOWER, &phy_data);
4886	if (!ret_val)
4887		adapter->stats.scc += phy_data;
4888
4889	/* Excessive Collision Count */
4890	hw->phy.ops.read_reg_page(hw, HV_ECOL_UPPER, &phy_data);
4891	ret_val = hw->phy.ops.read_reg_page(hw, HV_ECOL_LOWER, &phy_data);
4892	if (!ret_val)
4893		adapter->stats.ecol += phy_data;
4894
4895	/* Multiple Collision Count */
4896	hw->phy.ops.read_reg_page(hw, HV_MCC_UPPER, &phy_data);
4897	ret_val = hw->phy.ops.read_reg_page(hw, HV_MCC_LOWER, &phy_data);
4898	if (!ret_val)
4899		adapter->stats.mcc += phy_data;
4900
4901	/* Late Collision Count */
4902	hw->phy.ops.read_reg_page(hw, HV_LATECOL_UPPER, &phy_data);
4903	ret_val = hw->phy.ops.read_reg_page(hw, HV_LATECOL_LOWER, &phy_data);
4904	if (!ret_val)
4905		adapter->stats.latecol += phy_data;
4906
4907	/* Collision Count - also used for adaptive IFS */
4908	hw->phy.ops.read_reg_page(hw, HV_COLC_UPPER, &phy_data);
4909	ret_val = hw->phy.ops.read_reg_page(hw, HV_COLC_LOWER, &phy_data);
4910	if (!ret_val)
4911		hw->mac.collision_delta = phy_data;
4912
4913	/* Defer Count */
4914	hw->phy.ops.read_reg_page(hw, HV_DC_UPPER, &phy_data);
4915	ret_val = hw->phy.ops.read_reg_page(hw, HV_DC_LOWER, &phy_data);
4916	if (!ret_val)
4917		adapter->stats.dc += phy_data;
4918
4919	/* Transmit with no CRS */
4920	hw->phy.ops.read_reg_page(hw, HV_TNCRS_UPPER, &phy_data);
4921	ret_val = hw->phy.ops.read_reg_page(hw, HV_TNCRS_LOWER, &phy_data);
4922	if (!ret_val)
4923		adapter->stats.tncrs += phy_data;
4924
4925release:
4926	hw->phy.ops.release(hw);
4927}
4928
4929/**
4930 * e1000e_update_stats - Update the board statistics counters
4931 * @adapter: board private structure
4932 **/
4933static void e1000e_update_stats(struct e1000_adapter *adapter)
4934{
4935	struct net_device *netdev = adapter->netdev;
4936	struct e1000_hw *hw = &adapter->hw;
4937	struct pci_dev *pdev = adapter->pdev;
4938
4939	/* Prevent stats update while adapter is being reset, or if the pci
4940	 * connection is down.
4941	 */
4942	if (adapter->link_speed == 0)
4943		return;
4944	if (pci_channel_offline(pdev))
4945		return;
4946
4947	adapter->stats.crcerrs += er32(CRCERRS);
4948	adapter->stats.gprc += er32(GPRC);
4949	adapter->stats.gorc += er32(GORCL);
4950	er32(GORCH);		/* Clear gorc */
4951	adapter->stats.bprc += er32(BPRC);
4952	adapter->stats.mprc += er32(MPRC);
4953	adapter->stats.roc += er32(ROC);
4954
4955	adapter->stats.mpc += er32(MPC);
4956
4957	/* Half-duplex statistics */
4958	if (adapter->link_duplex == HALF_DUPLEX) {
4959		if (adapter->flags2 & FLAG2_HAS_PHY_STATS) {
4960			e1000e_update_phy_stats(adapter);
4961		} else {
4962			adapter->stats.scc += er32(SCC);
4963			adapter->stats.ecol += er32(ECOL);
4964			adapter->stats.mcc += er32(MCC);
4965			adapter->stats.latecol += er32(LATECOL);
4966			adapter->stats.dc += er32(DC);
4967
4968			hw->mac.collision_delta = er32(COLC);
4969
4970			if ((hw->mac.type != e1000_82574) &&
4971			    (hw->mac.type != e1000_82583))
4972				adapter->stats.tncrs += er32(TNCRS);
4973		}
4974		adapter->stats.colc += hw->mac.collision_delta;
4975	}
4976
4977	adapter->stats.xonrxc += er32(XONRXC);
4978	adapter->stats.xontxc += er32(XONTXC);
4979	adapter->stats.xoffrxc += er32(XOFFRXC);
4980	adapter->stats.xofftxc += er32(XOFFTXC);
4981	adapter->stats.gptc += er32(GPTC);
4982	adapter->stats.gotc += er32(GOTCL);
4983	er32(GOTCH);		/* Clear gotc */
4984	adapter->stats.rnbc += er32(RNBC);
4985	adapter->stats.ruc += er32(RUC);
4986
4987	adapter->stats.mptc += er32(MPTC);
4988	adapter->stats.bptc += er32(BPTC);
4989
4990	/* used for adaptive IFS */
4991
4992	hw->mac.tx_packet_delta = er32(TPT);
4993	adapter->stats.tpt += hw->mac.tx_packet_delta;
4994
4995	adapter->stats.algnerrc += er32(ALGNERRC);
4996	adapter->stats.rxerrc += er32(RXERRC);
4997	adapter->stats.cexterr += er32(CEXTERR);
4998	adapter->stats.tsctc += er32(TSCTC);
4999	adapter->stats.tsctfc += er32(TSCTFC);
5000
5001	/* Fill out the OS statistics structure */
5002	netdev->stats.multicast = adapter->stats.mprc;
5003	netdev->stats.collisions = adapter->stats.colc;
5004
5005	/* Rx Errors */
5006
5007	/* RLEC on some newer hardware can be incorrect so build
5008	 * our own version based on RUC and ROC
5009	 */
5010	netdev->stats.rx_errors = adapter->stats.rxerrc +
5011	    adapter->stats.crcerrs + adapter->stats.algnerrc +
5012	    adapter->stats.ruc + adapter->stats.roc + adapter->stats.cexterr;
5013	netdev->stats.rx_length_errors = adapter->stats.ruc +
5014	    adapter->stats.roc;
5015	netdev->stats.rx_crc_errors = adapter->stats.crcerrs;
5016	netdev->stats.rx_frame_errors = adapter->stats.algnerrc;
5017	netdev->stats.rx_missed_errors = adapter->stats.mpc;
5018
5019	/* Tx Errors */
5020	netdev->stats.tx_errors = adapter->stats.ecol + adapter->stats.latecol;
5021	netdev->stats.tx_aborted_errors = adapter->stats.ecol;
5022	netdev->stats.tx_window_errors = adapter->stats.latecol;
5023	netdev->stats.tx_carrier_errors = adapter->stats.tncrs;
5024
5025	/* Tx Dropped needs to be maintained elsewhere */
5026
5027	/* Management Stats */
5028	adapter->stats.mgptc += er32(MGTPTC);
5029	adapter->stats.mgprc += er32(MGTPRC);
5030	adapter->stats.mgpdc += er32(MGTPDC);
5031
5032	/* Correctable ECC Errors */
5033	if (hw->mac.type >= e1000_pch_lpt) {
5034		u32 pbeccsts = er32(PBECCSTS);
5035
5036		adapter->corr_errors +=
5037		    pbeccsts & E1000_PBECCSTS_CORR_ERR_CNT_MASK;
5038		adapter->uncorr_errors +=
5039		    (pbeccsts & E1000_PBECCSTS_UNCORR_ERR_CNT_MASK) >>
5040		    E1000_PBECCSTS_UNCORR_ERR_CNT_SHIFT;
5041	}
5042}
5043
5044/**
5045 * e1000_phy_read_status - Update the PHY register status snapshot
5046 * @adapter: board private structure
5047 **/
5048static void e1000_phy_read_status(struct e1000_adapter *adapter)
5049{
5050	struct e1000_hw *hw = &adapter->hw;
5051	struct e1000_phy_regs *phy = &adapter->phy_regs;
5052
5053	if (!pm_runtime_suspended((&adapter->pdev->dev)->parent) &&
5054	    (er32(STATUS) & E1000_STATUS_LU) &&
5055	    (adapter->hw.phy.media_type == e1000_media_type_copper)) {
5056		int ret_val;
5057
5058		ret_val = e1e_rphy(hw, MII_BMCR, &phy->bmcr);
5059		ret_val |= e1e_rphy(hw, MII_BMSR, &phy->bmsr);
5060		ret_val |= e1e_rphy(hw, MII_ADVERTISE, &phy->advertise);
5061		ret_val |= e1e_rphy(hw, MII_LPA, &phy->lpa);
5062		ret_val |= e1e_rphy(hw, MII_EXPANSION, &phy->expansion);
5063		ret_val |= e1e_rphy(hw, MII_CTRL1000, &phy->ctrl1000);
5064		ret_val |= e1e_rphy(hw, MII_STAT1000, &phy->stat1000);
5065		ret_val |= e1e_rphy(hw, MII_ESTATUS, &phy->estatus);
5066		if (ret_val)
5067			e_warn("Error reading PHY register\n");
5068	} else {
5069		/* Do not read PHY registers if link is not up
5070		 * Set values to typical power-on defaults
5071		 */
5072		phy->bmcr = (BMCR_SPEED1000 | BMCR_ANENABLE | BMCR_FULLDPLX);
5073		phy->bmsr = (BMSR_100FULL | BMSR_100HALF | BMSR_10FULL |
5074			     BMSR_10HALF | BMSR_ESTATEN | BMSR_ANEGCAPABLE |
5075			     BMSR_ERCAP);
5076		phy->advertise = (ADVERTISE_PAUSE_ASYM | ADVERTISE_PAUSE_CAP |
5077				  ADVERTISE_ALL | ADVERTISE_CSMA);
5078		phy->lpa = 0;
5079		phy->expansion = EXPANSION_ENABLENPAGE;
5080		phy->ctrl1000 = ADVERTISE_1000FULL;
5081		phy->stat1000 = 0;
5082		phy->estatus = (ESTATUS_1000_TFULL | ESTATUS_1000_THALF);
5083	}
5084}
5085
5086static void e1000_print_link_info(struct e1000_adapter *adapter)
5087{
5088	struct e1000_hw *hw = &adapter->hw;
5089	u32 ctrl = er32(CTRL);
5090
5091	/* Link status message must follow this format for user tools */
5092	netdev_info(adapter->netdev,
5093		    "NIC Link is Up %d Mbps %s Duplex, Flow Control: %s\n",
5094		    adapter->link_speed,
5095		    adapter->link_duplex == FULL_DUPLEX ? "Full" : "Half",
5096		    (ctrl & E1000_CTRL_TFCE) && (ctrl & E1000_CTRL_RFCE) ? "Rx/Tx" :
5097		    (ctrl & E1000_CTRL_RFCE) ? "Rx" :
5098		    (ctrl & E1000_CTRL_TFCE) ? "Tx" : "None");
5099}
5100
5101static bool e1000e_has_link(struct e1000_adapter *adapter)
5102{
5103	struct e1000_hw *hw = &adapter->hw;
5104	bool link_active = false;
5105	s32 ret_val = 0;
5106
5107	/* get_link_status is set on LSC (link status) interrupt or
5108	 * Rx sequence error interrupt.  get_link_status will stay
5109	 * true until the check_for_link establishes link
5110	 * for copper adapters ONLY
5111	 */
5112	switch (hw->phy.media_type) {
5113	case e1000_media_type_copper:
5114		if (hw->mac.get_link_status) {
5115			ret_val = hw->mac.ops.check_for_link(hw);
5116			link_active = !hw->mac.get_link_status;
5117		} else {
5118			link_active = true;
5119		}
5120		break;
5121	case e1000_media_type_fiber:
5122		ret_val = hw->mac.ops.check_for_link(hw);
5123		link_active = !!(er32(STATUS) & E1000_STATUS_LU);
5124		break;
5125	case e1000_media_type_internal_serdes:
5126		ret_val = hw->mac.ops.check_for_link(hw);
5127		link_active = hw->mac.serdes_has_link;
5128		break;
5129	default:
5130	case e1000_media_type_unknown:
5131		break;
5132	}
5133
5134	if ((ret_val == -E1000_ERR_PHY) && (hw->phy.type == e1000_phy_igp_3) &&
5135	    (er32(CTRL) & E1000_PHY_CTRL_GBE_DISABLE)) {
5136		/* See e1000_kmrn_lock_loss_workaround_ich8lan() */
5137		e_info("Gigabit has been disabled, downgrading speed\n");
5138	}
5139
5140	return link_active;
5141}
5142
5143static void e1000e_enable_receives(struct e1000_adapter *adapter)
5144{
5145	/* make sure the receive unit is started */
5146	if ((adapter->flags & FLAG_RX_NEEDS_RESTART) &&
5147	    (adapter->flags & FLAG_RESTART_NOW)) {
5148		struct e1000_hw *hw = &adapter->hw;
5149		u32 rctl = er32(RCTL);
5150
5151		ew32(RCTL, rctl | E1000_RCTL_EN);
5152		adapter->flags &= ~FLAG_RESTART_NOW;
5153	}
5154}
5155
5156static void e1000e_check_82574_phy_workaround(struct e1000_adapter *adapter)
5157{
5158	struct e1000_hw *hw = &adapter->hw;
5159
5160	/* With 82574 controllers, PHY needs to be checked periodically
5161	 * for hung state and reset, if two calls return true
5162	 */
5163	if (e1000_check_phy_82574(hw))
5164		adapter->phy_hang_count++;
5165	else
5166		adapter->phy_hang_count = 0;
5167
5168	if (adapter->phy_hang_count > 1) {
5169		adapter->phy_hang_count = 0;
5170		e_dbg("PHY appears hung - resetting\n");
5171		schedule_work(&adapter->reset_task);
5172	}
5173}
5174
5175/**
5176 * e1000_watchdog - Timer Call-back
5177 * @t: pointer to timer_list containing private info adapter
5178 **/
5179static void e1000_watchdog(struct timer_list *t)
5180{
5181	struct e1000_adapter *adapter = from_timer(adapter, t, watchdog_timer);
5182
5183	/* Do the rest outside of interrupt context */
5184	schedule_work(&adapter->watchdog_task);
5185
5186	/* TODO: make this use queue_delayed_work() */
5187}
5188
5189static void e1000_watchdog_task(struct work_struct *work)
5190{
5191	struct e1000_adapter *adapter = container_of(work,
5192						     struct e1000_adapter,
5193						     watchdog_task);
5194	struct net_device *netdev = adapter->netdev;
5195	struct e1000_mac_info *mac = &adapter->hw.mac;
5196	struct e1000_phy_info *phy = &adapter->hw.phy;
5197	struct e1000_ring *tx_ring = adapter->tx_ring;
5198	u32 dmoff_exit_timeout = 100, tries = 0;
5199	struct e1000_hw *hw = &adapter->hw;
5200	u32 link, tctl, pcim_state;
5201
5202	if (test_bit(__E1000_DOWN, &adapter->state))
5203		return;
5204
5205	link = e1000e_has_link(adapter);
5206	if ((netif_carrier_ok(netdev)) && link) {
5207		/* Cancel scheduled suspend requests. */
5208		pm_runtime_resume(netdev->dev.parent);
5209
5210		e1000e_enable_receives(adapter);
5211		goto link_up;
5212	}
5213
5214	if ((e1000e_enable_tx_pkt_filtering(hw)) &&
5215	    (adapter->mng_vlan_id != adapter->hw.mng_cookie.vlan_id))
5216		e1000_update_mng_vlan(adapter);
5217
5218	if (link) {
5219		if (!netif_carrier_ok(netdev)) {
5220			bool txb2b = true;
5221
5222			/* Cancel scheduled suspend requests. */
5223			pm_runtime_resume(netdev->dev.parent);
5224
5225			/* Checking if MAC is in DMoff state*/
5226			if (er32(FWSM) & E1000_ICH_FWSM_FW_VALID) {
5227				pcim_state = er32(STATUS);
5228				while (pcim_state & E1000_STATUS_PCIM_STATE) {
5229					if (tries++ == dmoff_exit_timeout) {
5230						e_dbg("Error in exiting dmoff\n");
5231						break;
5232					}
5233					usleep_range(10000, 20000);
5234					pcim_state = er32(STATUS);
5235
5236					/* Checking if MAC exited DMoff state */
5237					if (!(pcim_state & E1000_STATUS_PCIM_STATE))
5238						e1000_phy_hw_reset(&adapter->hw);
5239				}
5240			}
5241
5242			/* update snapshot of PHY registers on LSC */
5243			e1000_phy_read_status(adapter);
5244			mac->ops.get_link_up_info(&adapter->hw,
5245						  &adapter->link_speed,
5246						  &adapter->link_duplex);
5247			e1000_print_link_info(adapter);
5248
5249			/* check if SmartSpeed worked */
5250			e1000e_check_downshift(hw);
5251			if (phy->speed_downgraded)
5252				netdev_warn(netdev,
5253					    "Link Speed was downgraded by SmartSpeed\n");
5254
5255			/* On supported PHYs, check for duplex mismatch only
5256			 * if link has autonegotiated at 10/100 half
5257			 */
5258			if ((hw->phy.type == e1000_phy_igp_3 ||
5259			     hw->phy.type == e1000_phy_bm) &&
5260			    hw->mac.autoneg &&
5261			    (adapter->link_speed == SPEED_10 ||
5262			     adapter->link_speed == SPEED_100) &&
5263			    (adapter->link_duplex == HALF_DUPLEX)) {
5264				u16 autoneg_exp;
5265
5266				e1e_rphy(hw, MII_EXPANSION, &autoneg_exp);
5267
5268				if (!(autoneg_exp & EXPANSION_NWAY))
5269					e_info("Autonegotiated half duplex but link partner cannot autoneg.  Try forcing full duplex if link gets many collisions.\n");
5270			}
5271
5272			/* adjust timeout factor according to speed/duplex */
5273			adapter->tx_timeout_factor = 1;
5274			switch (adapter->link_speed) {
5275			case SPEED_10:
5276				txb2b = false;
5277				adapter->tx_timeout_factor = 16;
5278				break;
5279			case SPEED_100:
5280				txb2b = false;
5281				adapter->tx_timeout_factor = 10;
5282				break;
5283			}
5284
5285			/* workaround: re-program speed mode bit after
5286			 * link-up event
5287			 */
5288			if ((adapter->flags & FLAG_TARC_SPEED_MODE_BIT) &&
5289			    !txb2b) {
5290				u32 tarc0;
5291
5292				tarc0 = er32(TARC(0));
5293				tarc0 &= ~SPEED_MODE_BIT;
5294				ew32(TARC(0), tarc0);
5295			}
5296
5297			/* enable transmits in the hardware, need to do this
5298			 * after setting TARC(0)
5299			 */
5300			tctl = er32(TCTL);
5301			tctl |= E1000_TCTL_EN;
5302			ew32(TCTL, tctl);
5303
5304			/* Perform any post-link-up configuration before
5305			 * reporting link up.
5306			 */
5307			if (phy->ops.cfg_on_link_up)
5308				phy->ops.cfg_on_link_up(hw);
5309
5310			netif_wake_queue(netdev);
5311			netif_carrier_on(netdev);
5312
5313			if (!test_bit(__E1000_DOWN, &adapter->state))
5314				mod_timer(&adapter->phy_info_timer,
5315					  round_jiffies(jiffies + 2 * HZ));
5316		}
5317	} else {
5318		if (netif_carrier_ok(netdev)) {
5319			adapter->link_speed = 0;
5320			adapter->link_duplex = 0;
5321			/* Link status message must follow this format */
5322			netdev_info(netdev, "NIC Link is Down\n");
5323			netif_carrier_off(netdev);
5324			netif_stop_queue(netdev);
5325			if (!test_bit(__E1000_DOWN, &adapter->state))
5326				mod_timer(&adapter->phy_info_timer,
5327					  round_jiffies(jiffies + 2 * HZ));
5328
5329			/* 8000ES2LAN requires a Rx packet buffer work-around
5330			 * on link down event; reset the controller to flush
5331			 * the Rx packet buffer.
5332			 */
5333			if (adapter->flags & FLAG_RX_NEEDS_RESTART)
5334				adapter->flags |= FLAG_RESTART_NOW;
5335			else
5336				pm_schedule_suspend(netdev->dev.parent,
5337						    LINK_TIMEOUT);
5338		}
5339	}
5340
5341link_up:
5342	spin_lock(&adapter->stats64_lock);
5343	e1000e_update_stats(adapter);
5344
5345	mac->tx_packet_delta = adapter->stats.tpt - adapter->tpt_old;
5346	adapter->tpt_old = adapter->stats.tpt;
5347	mac->collision_delta = adapter->stats.colc - adapter->colc_old;
5348	adapter->colc_old = adapter->stats.colc;
5349
5350	adapter->gorc = adapter->stats.gorc - adapter->gorc_old;
5351	adapter->gorc_old = adapter->stats.gorc;
5352	adapter->gotc = adapter->stats.gotc - adapter->gotc_old;
5353	adapter->gotc_old = adapter->stats.gotc;
5354	spin_unlock(&adapter->stats64_lock);
5355
5356	/* If the link is lost the controller stops DMA, but
5357	 * if there is queued Tx work it cannot be done.  So
5358	 * reset the controller to flush the Tx packet buffers.
5359	 */
5360	if (!netif_carrier_ok(netdev) &&
5361	    (e1000_desc_unused(tx_ring) + 1 < tx_ring->count))
5362		adapter->flags |= FLAG_RESTART_NOW;
5363
5364	/* If reset is necessary, do it outside of interrupt context. */
5365	if (adapter->flags & FLAG_RESTART_NOW) {
5366		schedule_work(&adapter->reset_task);
5367		/* return immediately since reset is imminent */
5368		return;
5369	}
5370
5371	e1000e_update_adaptive(&adapter->hw);
5372
5373	/* Simple mode for Interrupt Throttle Rate (ITR) */
5374	if (adapter->itr_setting == 4) {
5375		/* Symmetric Tx/Rx gets a reduced ITR=2000;
5376		 * Total asymmetrical Tx or Rx gets ITR=8000;
5377		 * everyone else is between 2000-8000.
5378		 */
5379		u32 goc = (adapter->gotc + adapter->gorc) / 10000;
5380		u32 dif = (adapter->gotc > adapter->gorc ?
5381			   adapter->gotc - adapter->gorc :
5382			   adapter->gorc - adapter->gotc) / 10000;
5383		u32 itr = goc > 0 ? (dif * 6000 / goc + 2000) : 8000;
5384
5385		e1000e_write_itr(adapter, itr);
5386	}
5387
5388	/* Cause software interrupt to ensure Rx ring is cleaned */
5389	if (adapter->msix_entries)
5390		ew32(ICS, adapter->rx_ring->ims_val);
5391	else
5392		ew32(ICS, E1000_ICS_RXDMT0);
5393
5394	/* flush pending descriptors to memory before detecting Tx hang */
5395	e1000e_flush_descriptors(adapter);
5396
5397	/* Force detection of hung controller every watchdog period */
5398	adapter->detect_tx_hung = true;
5399
5400	/* With 82571 controllers, LAA may be overwritten due to controller
5401	 * reset from the other port. Set the appropriate LAA in RAR[0]
5402	 */
5403	if (e1000e_get_laa_state_82571(hw))
5404		hw->mac.ops.rar_set(hw, adapter->hw.mac.addr, 0);
5405
5406	if (adapter->flags2 & FLAG2_CHECK_PHY_HANG)
5407		e1000e_check_82574_phy_workaround(adapter);
5408
5409	/* Clear valid timestamp stuck in RXSTMPL/H due to a Rx error */
5410	if (adapter->hwtstamp_config.rx_filter != HWTSTAMP_FILTER_NONE) {
5411		if ((adapter->flags2 & FLAG2_CHECK_RX_HWTSTAMP) &&
5412		    (er32(TSYNCRXCTL) & E1000_TSYNCRXCTL_VALID)) {
5413			er32(RXSTMPH);
5414			adapter->rx_hwtstamp_cleared++;
5415		} else {
5416			adapter->flags2 |= FLAG2_CHECK_RX_HWTSTAMP;
5417		}
5418	}
5419
5420	/* Reset the timer */
5421	if (!test_bit(__E1000_DOWN, &adapter->state))
5422		mod_timer(&adapter->watchdog_timer,
5423			  round_jiffies(jiffies + 2 * HZ));
5424}
5425
5426#define E1000_TX_FLAGS_CSUM		0x00000001
5427#define E1000_TX_FLAGS_VLAN		0x00000002
5428#define E1000_TX_FLAGS_TSO		0x00000004
5429#define E1000_TX_FLAGS_IPV4		0x00000008
5430#define E1000_TX_FLAGS_NO_FCS		0x00000010
5431#define E1000_TX_FLAGS_HWTSTAMP		0x00000020
5432#define E1000_TX_FLAGS_VLAN_MASK	0xffff0000
5433#define E1000_TX_FLAGS_VLAN_SHIFT	16
5434
5435static int e1000_tso(struct e1000_ring *tx_ring, struct sk_buff *skb,
5436		     __be16 protocol)
5437{
5438	struct e1000_context_desc *context_desc;
5439	struct e1000_buffer *buffer_info;
5440	unsigned int i;
5441	u32 cmd_length = 0;
5442	u16 ipcse = 0, mss;
5443	u8 ipcss, ipcso, tucss, tucso, hdr_len;
5444	int err;
5445
5446	if (!skb_is_gso(skb))
5447		return 0;
5448
5449	err = skb_cow_head(skb, 0);
5450	if (err < 0)
5451		return err;
5452
5453	hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
5454	mss = skb_shinfo(skb)->gso_size;
5455	if (protocol == htons(ETH_P_IP)) {
5456		struct iphdr *iph = ip_hdr(skb);
5457		iph->tot_len = 0;
5458		iph->check = 0;
5459		tcp_hdr(skb)->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr,
5460							 0, IPPROTO_TCP, 0);
5461		cmd_length = E1000_TXD_CMD_IP;
5462		ipcse = skb_transport_offset(skb) - 1;
5463	} else if (skb_is_gso_v6(skb)) {
5464		tcp_v6_gso_csum_prep(skb);
5465		ipcse = 0;
5466	}
5467	ipcss = skb_network_offset(skb);
5468	ipcso = (void *)&(ip_hdr(skb)->check) - (void *)skb->data;
5469	tucss = skb_transport_offset(skb);
5470	tucso = (void *)&(tcp_hdr(skb)->check) - (void *)skb->data;
5471
5472	cmd_length |= (E1000_TXD_CMD_DEXT | E1000_TXD_CMD_TSE |
5473		       E1000_TXD_CMD_TCP | (skb->len - (hdr_len)));
5474
5475	i = tx_ring->next_to_use;
5476	context_desc = E1000_CONTEXT_DESC(*tx_ring, i);
5477	buffer_info = &tx_ring->buffer_info[i];
5478
5479	context_desc->lower_setup.ip_fields.ipcss = ipcss;
5480	context_desc->lower_setup.ip_fields.ipcso = ipcso;
5481	context_desc->lower_setup.ip_fields.ipcse = cpu_to_le16(ipcse);
5482	context_desc->upper_setup.tcp_fields.tucss = tucss;
5483	context_desc->upper_setup.tcp_fields.tucso = tucso;
5484	context_desc->upper_setup.tcp_fields.tucse = 0;
5485	context_desc->tcp_seg_setup.fields.mss = cpu_to_le16(mss);
5486	context_desc->tcp_seg_setup.fields.hdr_len = hdr_len;
5487	context_desc->cmd_and_length = cpu_to_le32(cmd_length);
5488
5489	buffer_info->time_stamp = jiffies;
5490	buffer_info->next_to_watch = i;
5491
5492	i++;
5493	if (i == tx_ring->count)
5494		i = 0;
5495	tx_ring->next_to_use = i;
5496
5497	return 1;
5498}
5499
5500static bool e1000_tx_csum(struct e1000_ring *tx_ring, struct sk_buff *skb,
5501			  __be16 protocol)
5502{
5503	struct e1000_adapter *adapter = tx_ring->adapter;
5504	struct e1000_context_desc *context_desc;
5505	struct e1000_buffer *buffer_info;
5506	unsigned int i;
5507	u8 css;
5508	u32 cmd_len = E1000_TXD_CMD_DEXT;
5509
5510	if (skb->ip_summed != CHECKSUM_PARTIAL)
5511		return false;
5512
5513	switch (protocol) {
5514	case cpu_to_be16(ETH_P_IP):
5515		if (ip_hdr(skb)->protocol == IPPROTO_TCP)
5516			cmd_len |= E1000_TXD_CMD_TCP;
5517		break;
5518	case cpu_to_be16(ETH_P_IPV6):
5519		/* XXX not handling all IPV6 headers */
5520		if (ipv6_hdr(skb)->nexthdr == IPPROTO_TCP)
5521			cmd_len |= E1000_TXD_CMD_TCP;
5522		break;
5523	default:
5524		if (unlikely(net_ratelimit()))
5525			e_warn("checksum_partial proto=%x!\n",
5526			       be16_to_cpu(protocol));
5527		break;
5528	}
5529
5530	css = skb_checksum_start_offset(skb);
5531
5532	i = tx_ring->next_to_use;
5533	buffer_info = &tx_ring->buffer_info[i];
5534	context_desc = E1000_CONTEXT_DESC(*tx_ring, i);
5535
5536	context_desc->lower_setup.ip_config = 0;
5537	context_desc->upper_setup.tcp_fields.tucss = css;
5538	context_desc->upper_setup.tcp_fields.tucso = css + skb->csum_offset;
5539	context_desc->upper_setup.tcp_fields.tucse = 0;
5540	context_desc->tcp_seg_setup.data = 0;
5541	context_desc->cmd_and_length = cpu_to_le32(cmd_len);
5542
5543	buffer_info->time_stamp = jiffies;
5544	buffer_info->next_to_watch = i;
5545
5546	i++;
5547	if (i == tx_ring->count)
5548		i = 0;
5549	tx_ring->next_to_use = i;
5550
5551	return true;
5552}
5553
5554static int e1000_tx_map(struct e1000_ring *tx_ring, struct sk_buff *skb,
5555			unsigned int first, unsigned int max_per_txd,
5556			unsigned int nr_frags)
5557{
5558	struct e1000_adapter *adapter = tx_ring->adapter;
5559	struct pci_dev *pdev = adapter->pdev;
5560	struct e1000_buffer *buffer_info;
5561	unsigned int len = skb_headlen(skb);
5562	unsigned int offset = 0, size, count = 0, i;
5563	unsigned int f, bytecount, segs;
5564
5565	i = tx_ring->next_to_use;
5566
5567	while (len) {
5568		buffer_info = &tx_ring->buffer_info[i];
5569		size = min(len, max_per_txd);
5570
5571		buffer_info->length = size;
5572		buffer_info->time_stamp = jiffies;
5573		buffer_info->next_to_watch = i;
5574		buffer_info->dma = dma_map_single(&pdev->dev,
5575						  skb->data + offset,
5576						  size, DMA_TO_DEVICE);
5577		buffer_info->mapped_as_page = false;
5578		if (dma_mapping_error(&pdev->dev, buffer_info->dma))
5579			goto dma_error;
5580
5581		len -= size;
5582		offset += size;
5583		count++;
5584
5585		if (len) {
5586			i++;
5587			if (i == tx_ring->count)
5588				i = 0;
5589		}
5590	}
5591
5592	for (f = 0; f < nr_frags; f++) {
5593		const skb_frag_t *frag = &skb_shinfo(skb)->frags[f];
5594
5595		len = skb_frag_size(frag);
5596		offset = 0;
5597
5598		while (len) {
5599			i++;
5600			if (i == tx_ring->count)
5601				i = 0;
5602
5603			buffer_info = &tx_ring->buffer_info[i];
5604			size = min(len, max_per_txd);
5605
5606			buffer_info->length = size;
5607			buffer_info->time_stamp = jiffies;
5608			buffer_info->next_to_watch = i;
5609			buffer_info->dma = skb_frag_dma_map(&pdev->dev, frag,
5610							    offset, size,
5611							    DMA_TO_DEVICE);
5612			buffer_info->mapped_as_page = true;
5613			if (dma_mapping_error(&pdev->dev, buffer_info->dma))
5614				goto dma_error;
5615
5616			len -= size;
5617			offset += size;
5618			count++;
5619		}
5620	}
5621
5622	segs = skb_shinfo(skb)->gso_segs ? : 1;
5623	/* multiply data chunks by size of headers */
5624	bytecount = ((segs - 1) * skb_headlen(skb)) + skb->len;
5625
5626	tx_ring->buffer_info[i].skb = skb;
5627	tx_ring->buffer_info[i].segs = segs;
5628	tx_ring->buffer_info[i].bytecount = bytecount;
5629	tx_ring->buffer_info[first].next_to_watch = i;
5630
5631	return count;
5632
5633dma_error:
5634	dev_err(&pdev->dev, "Tx DMA map failed\n");
5635	buffer_info->dma = 0;
5636	if (count)
5637		count--;
5638
5639	while (count--) {
5640		if (i == 0)
5641			i += tx_ring->count;
5642		i--;
5643		buffer_info = &tx_ring->buffer_info[i];
5644		e1000_put_txbuf(tx_ring, buffer_info, true);
5645	}
5646
5647	return 0;
5648}
5649
5650static void e1000_tx_queue(struct e1000_ring *tx_ring, int tx_flags, int count)
5651{
5652	struct e1000_adapter *adapter = tx_ring->adapter;
5653	struct e1000_tx_desc *tx_desc = NULL;
5654	struct e1000_buffer *buffer_info;
5655	u32 txd_upper = 0, txd_lower = E1000_TXD_CMD_IFCS;
5656	unsigned int i;
5657
5658	if (tx_flags & E1000_TX_FLAGS_TSO) {
5659		txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D |
5660		    E1000_TXD_CMD_TSE;
5661		txd_upper |= E1000_TXD_POPTS_TXSM << 8;
5662
5663		if (tx_flags & E1000_TX_FLAGS_IPV4)
5664			txd_upper |= E1000_TXD_POPTS_IXSM << 8;
5665	}
5666
5667	if (tx_flags & E1000_TX_FLAGS_CSUM) {
5668		txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D;
5669		txd_upper |= E1000_TXD_POPTS_TXSM << 8;
5670	}
5671
5672	if (tx_flags & E1000_TX_FLAGS_VLAN) {
5673		txd_lower |= E1000_TXD_CMD_VLE;
5674		txd_upper |= (tx_flags & E1000_TX_FLAGS_VLAN_MASK);
5675	}
5676
5677	if (unlikely(tx_flags & E1000_TX_FLAGS_NO_FCS))
5678		txd_lower &= ~(E1000_TXD_CMD_IFCS);
5679
5680	if (unlikely(tx_flags & E1000_TX_FLAGS_HWTSTAMP)) {
5681		txd_lower |= E1000_TXD_CMD_DEXT | E1000_TXD_DTYP_D;
5682		txd_upper |= E1000_TXD_EXTCMD_TSTAMP;
5683	}
5684
5685	i = tx_ring->next_to_use;
5686
5687	do {
5688		buffer_info = &tx_ring->buffer_info[i];
5689		tx_desc = E1000_TX_DESC(*tx_ring, i);
5690		tx_desc->buffer_addr = cpu_to_le64(buffer_info->dma);
5691		tx_desc->lower.data = cpu_to_le32(txd_lower |
5692						  buffer_info->length);
5693		tx_desc->upper.data = cpu_to_le32(txd_upper);
5694
5695		i++;
5696		if (i == tx_ring->count)
5697			i = 0;
5698	} while (--count > 0);
5699
5700	tx_desc->lower.data |= cpu_to_le32(adapter->txd_cmd);
5701
5702	/* txd_cmd re-enables FCS, so we'll re-disable it here as desired. */
5703	if (unlikely(tx_flags & E1000_TX_FLAGS_NO_FCS))
5704		tx_desc->lower.data &= ~(cpu_to_le32(E1000_TXD_CMD_IFCS));
5705
5706	/* Force memory writes to complete before letting h/w
5707	 * know there are new descriptors to fetch.  (Only
5708	 * applicable for weak-ordered memory model archs,
5709	 * such as IA-64).
5710	 */
5711	wmb();
5712
5713	tx_ring->next_to_use = i;
5714}
5715
5716#define MINIMUM_DHCP_PACKET_SIZE 282
5717static int e1000_transfer_dhcp_info(struct e1000_adapter *adapter,
5718				    struct sk_buff *skb)
5719{
5720	struct e1000_hw *hw = &adapter->hw;
5721	u16 length, offset;
5722
5723	if (skb_vlan_tag_present(skb) &&
5724	    !((skb_vlan_tag_get(skb) == adapter->hw.mng_cookie.vlan_id) &&
5725	      (adapter->hw.mng_cookie.status &
5726	       E1000_MNG_DHCP_COOKIE_STATUS_VLAN)))
5727		return 0;
5728
5729	if (skb->len <= MINIMUM_DHCP_PACKET_SIZE)
5730		return 0;
5731
5732	if (((struct ethhdr *)skb->data)->h_proto != htons(ETH_P_IP))
5733		return 0;
5734
5735	{
5736		const struct iphdr *ip = (struct iphdr *)((u8 *)skb->data + 14);
5737		struct udphdr *udp;
5738
5739		if (ip->protocol != IPPROTO_UDP)
5740			return 0;
5741
5742		udp = (struct udphdr *)((u8 *)ip + (ip->ihl << 2));
5743		if (ntohs(udp->dest) != 67)
5744			return 0;
5745
5746		offset = (u8 *)udp + 8 - skb->data;
5747		length = skb->len - offset;
5748		return e1000e_mng_write_dhcp_info(hw, (u8 *)udp + 8, length);
5749	}
5750
5751	return 0;
5752}
5753
5754static int __e1000_maybe_stop_tx(struct e1000_ring *tx_ring, int size)
5755{
5756	struct e1000_adapter *adapter = tx_ring->adapter;
5757
5758	netif_stop_queue(adapter->netdev);
5759	/* Herbert's original patch had:
5760	 *  smp_mb__after_netif_stop_queue();
5761	 * but since that doesn't exist yet, just open code it.
5762	 */
5763	smp_mb();
5764
5765	/* We need to check again in a case another CPU has just
5766	 * made room available.
5767	 */
5768	if (e1000_desc_unused(tx_ring) < size)
5769		return -EBUSY;
5770
5771	/* A reprieve! */
5772	netif_start_queue(adapter->netdev);
5773	++adapter->restart_queue;
5774	return 0;
5775}
5776
5777static int e1000_maybe_stop_tx(struct e1000_ring *tx_ring, int size)
5778{
5779	BUG_ON(size > tx_ring->count);
5780
5781	if (e1000_desc_unused(tx_ring) >= size)
5782		return 0;
5783	return __e1000_maybe_stop_tx(tx_ring, size);
5784}
5785
5786static netdev_tx_t e1000_xmit_frame(struct sk_buff *skb,
5787				    struct net_device *netdev)
5788{
5789	struct e1000_adapter *adapter = netdev_priv(netdev);
5790	struct e1000_ring *tx_ring = adapter->tx_ring;
5791	unsigned int first;
5792	unsigned int tx_flags = 0;
5793	unsigned int len = skb_headlen(skb);
5794	unsigned int nr_frags;
5795	unsigned int mss;
5796	int count = 0;
5797	int tso;
5798	unsigned int f;
5799	__be16 protocol = vlan_get_protocol(skb);
5800
5801	if (test_bit(__E1000_DOWN, &adapter->state)) {
5802		dev_kfree_skb_any(skb);
5803		return NETDEV_TX_OK;
5804	}
5805
5806	if (skb->len <= 0) {
5807		dev_kfree_skb_any(skb);
5808		return NETDEV_TX_OK;
5809	}
5810
5811	/* The minimum packet size with TCTL.PSP set is 17 bytes so
5812	 * pad skb in order to meet this minimum size requirement
5813	 */
5814	if (skb_put_padto(skb, 17))
5815		return NETDEV_TX_OK;
5816
5817	mss = skb_shinfo(skb)->gso_size;
5818	if (mss) {
5819		u8 hdr_len;
5820
5821		/* TSO Workaround for 82571/2/3 Controllers -- if skb->data
5822		 * points to just header, pull a few bytes of payload from
5823		 * frags into skb->data
5824		 */
5825		hdr_len = skb_transport_offset(skb) + tcp_hdrlen(skb);
5826		/* we do this workaround for ES2LAN, but it is un-necessary,
5827		 * avoiding it could save a lot of cycles
5828		 */
5829		if (skb->data_len && (hdr_len == len)) {
5830			unsigned int pull_size;
5831
5832			pull_size = min_t(unsigned int, 4, skb->data_len);
5833			if (!__pskb_pull_tail(skb, pull_size)) {
5834				e_err("__pskb_pull_tail failed.\n");
5835				dev_kfree_skb_any(skb);
5836				return NETDEV_TX_OK;
5837			}
5838			len = skb_headlen(skb);
5839		}
5840	}
5841
5842	/* reserve a descriptor for the offload context */
5843	if ((mss) || (skb->ip_summed == CHECKSUM_PARTIAL))
5844		count++;
5845	count++;
5846
5847	count += DIV_ROUND_UP(len, adapter->tx_fifo_limit);
5848
5849	nr_frags = skb_shinfo(skb)->nr_frags;
5850	for (f = 0; f < nr_frags; f++)
5851		count += DIV_ROUND_UP(skb_frag_size(&skb_shinfo(skb)->frags[f]),
5852				      adapter->tx_fifo_limit);
5853
5854	if (adapter->hw.mac.tx_pkt_filtering)
5855		e1000_transfer_dhcp_info(adapter, skb);
5856
5857	/* need: count + 2 desc gap to keep tail from touching
5858	 * head, otherwise try next time
5859	 */
5860	if (e1000_maybe_stop_tx(tx_ring, count + 2))
5861		return NETDEV_TX_BUSY;
5862
5863	if (skb_vlan_tag_present(skb)) {
5864		tx_flags |= E1000_TX_FLAGS_VLAN;
5865		tx_flags |= (skb_vlan_tag_get(skb) <<
5866			     E1000_TX_FLAGS_VLAN_SHIFT);
5867	}
5868
5869	first = tx_ring->next_to_use;
5870
5871	tso = e1000_tso(tx_ring, skb, protocol);
5872	if (tso < 0) {
5873		dev_kfree_skb_any(skb);
5874		return NETDEV_TX_OK;
5875	}
5876
5877	if (tso)
5878		tx_flags |= E1000_TX_FLAGS_TSO;
5879	else if (e1000_tx_csum(tx_ring, skb, protocol))
5880		tx_flags |= E1000_TX_FLAGS_CSUM;
5881
5882	/* Old method was to assume IPv4 packet by default if TSO was enabled.
5883	 * 82571 hardware supports TSO capabilities for IPv6 as well...
5884	 * no longer assume, we must.
5885	 */
5886	if (protocol == htons(ETH_P_IP))
5887		tx_flags |= E1000_TX_FLAGS_IPV4;
5888
5889	if (unlikely(skb->no_fcs))
5890		tx_flags |= E1000_TX_FLAGS_NO_FCS;
5891
5892	/* if count is 0 then mapping error has occurred */
5893	count = e1000_tx_map(tx_ring, skb, first, adapter->tx_fifo_limit,
5894			     nr_frags);
5895	if (count) {
5896		if (unlikely(skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP) &&
5897		    (adapter->flags & FLAG_HAS_HW_TIMESTAMP)) {
5898			if (!adapter->tx_hwtstamp_skb) {
5899				skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
5900				tx_flags |= E1000_TX_FLAGS_HWTSTAMP;
5901				adapter->tx_hwtstamp_skb = skb_get(skb);
5902				adapter->tx_hwtstamp_start = jiffies;
5903				schedule_work(&adapter->tx_hwtstamp_work);
5904			} else {
5905				adapter->tx_hwtstamp_skipped++;
5906			}
5907		}
5908
5909		skb_tx_timestamp(skb);
5910
5911		netdev_sent_queue(netdev, skb->len);
5912		e1000_tx_queue(tx_ring, tx_flags, count);
5913		/* Make sure there is space in the ring for the next send. */
5914		e1000_maybe_stop_tx(tx_ring,
5915				    ((MAX_SKB_FRAGS + 1) *
5916				     DIV_ROUND_UP(PAGE_SIZE,
5917						  adapter->tx_fifo_limit) + 4));
5918
5919		if (!netdev_xmit_more() ||
5920		    netif_xmit_stopped(netdev_get_tx_queue(netdev, 0))) {
5921			if (adapter->flags2 & FLAG2_PCIM2PCI_ARBITER_WA)
5922				e1000e_update_tdt_wa(tx_ring,
5923						     tx_ring->next_to_use);
5924			else
5925				writel(tx_ring->next_to_use, tx_ring->tail);
5926		}
5927	} else {
5928		dev_kfree_skb_any(skb);
5929		tx_ring->buffer_info[first].time_stamp = 0;
5930		tx_ring->next_to_use = first;
5931	}
5932
5933	return NETDEV_TX_OK;
5934}
5935
5936/**
5937 * e1000_tx_timeout - Respond to a Tx Hang
5938 * @netdev: network interface device structure
5939 * @txqueue: index of the hung queue (unused)
5940 **/
5941static void e1000_tx_timeout(struct net_device *netdev, unsigned int __always_unused txqueue)
5942{
5943	struct e1000_adapter *adapter = netdev_priv(netdev);
5944
5945	/* Do the reset outside of interrupt context */
5946	adapter->tx_timeout_count++;
5947	schedule_work(&adapter->reset_task);
5948}
5949
5950static void e1000_reset_task(struct work_struct *work)
5951{
5952	struct e1000_adapter *adapter;
5953	adapter = container_of(work, struct e1000_adapter, reset_task);
5954
5955	rtnl_lock();
5956	/* don't run the task if already down */
5957	if (test_bit(__E1000_DOWN, &adapter->state)) {
5958		rtnl_unlock();
5959		return;
5960	}
5961
5962	if (!(adapter->flags & FLAG_RESTART_NOW)) {
5963		e1000e_dump(adapter);
5964		e_err("Reset adapter unexpectedly\n");
5965	}
5966	e1000e_reinit_locked(adapter);
5967	rtnl_unlock();
5968}
5969
5970/**
5971 * e1000_get_stats64 - Get System Network Statistics
5972 * @netdev: network interface device structure
5973 * @stats: rtnl_link_stats64 pointer
5974 *
5975 * Returns the address of the device statistics structure.
5976 **/
5977void e1000e_get_stats64(struct net_device *netdev,
5978			struct rtnl_link_stats64 *stats)
5979{
5980	struct e1000_adapter *adapter = netdev_priv(netdev);
5981
5982	spin_lock(&adapter->stats64_lock);
5983	e1000e_update_stats(adapter);
5984	/* Fill out the OS statistics structure */
5985	stats->rx_bytes = adapter->stats.gorc;
5986	stats->rx_packets = adapter->stats.gprc;
5987	stats->tx_bytes = adapter->stats.gotc;
5988	stats->tx_packets = adapter->stats.gptc;
5989	stats->multicast = adapter->stats.mprc;
5990	stats->collisions = adapter->stats.colc;
5991
5992	/* Rx Errors */
5993
5994	/* RLEC on some newer hardware can be incorrect so build
5995	 * our own version based on RUC and ROC
5996	 */
5997	stats->rx_errors = adapter->stats.rxerrc +
5998	    adapter->stats.crcerrs + adapter->stats.algnerrc +
5999	    adapter->stats.ruc + adapter->stats.roc + adapter->stats.cexterr;
6000	stats->rx_length_errors = adapter->stats.ruc + adapter->stats.roc;
6001	stats->rx_crc_errors = adapter->stats.crcerrs;
6002	stats->rx_frame_errors = adapter->stats.algnerrc;
6003	stats->rx_missed_errors = adapter->stats.mpc;
6004
6005	/* Tx Errors */
6006	stats->tx_errors = adapter->stats.ecol + adapter->stats.latecol;
6007	stats->tx_aborted_errors = adapter->stats.ecol;
6008	stats->tx_window_errors = adapter->stats.latecol;
6009	stats->tx_carrier_errors = adapter->stats.tncrs;
6010
6011	/* Tx Dropped needs to be maintained elsewhere */
6012
6013	spin_unlock(&adapter->stats64_lock);
6014}
6015
6016/**
6017 * e1000_change_mtu - Change the Maximum Transfer Unit
6018 * @netdev: network interface device structure
6019 * @new_mtu: new value for maximum frame size
6020 *
6021 * Returns 0 on success, negative on failure
6022 **/
6023static int e1000_change_mtu(struct net_device *netdev, int new_mtu)
6024{
6025	struct e1000_adapter *adapter = netdev_priv(netdev);
6026	int max_frame = new_mtu + VLAN_ETH_HLEN + ETH_FCS_LEN;
6027
6028	/* Jumbo frame support */
6029	if ((new_mtu > ETH_DATA_LEN) &&
6030	    !(adapter->flags & FLAG_HAS_JUMBO_FRAMES)) {
6031		e_err("Jumbo Frames not supported.\n");
6032		return -EINVAL;
6033	}
6034
6035	/* Jumbo frame workaround on 82579 and newer requires CRC be stripped */
6036	if ((adapter->hw.mac.type >= e1000_pch2lan) &&
6037	    !(adapter->flags2 & FLAG2_CRC_STRIPPING) &&
6038	    (new_mtu > ETH_DATA_LEN)) {
6039		e_err("Jumbo Frames not supported on this device when CRC stripping is disabled.\n");
6040		return -EINVAL;
6041	}
6042
6043	while (test_and_set_bit(__E1000_RESETTING, &adapter->state))
6044		usleep_range(1000, 1100);
6045	/* e1000e_down -> e1000e_reset dependent on max_frame_size & mtu */
6046	adapter->max_frame_size = max_frame;
6047	netdev_dbg(netdev, "changing MTU from %d to %d\n",
6048		   netdev->mtu, new_mtu);
6049	netdev->mtu = new_mtu;
6050
6051	pm_runtime_get_sync(netdev->dev.parent);
6052
6053	if (netif_running(netdev))
6054		e1000e_down(adapter, true);
6055
6056	/* NOTE: netdev_alloc_skb reserves 16 bytes, and typically NET_IP_ALIGN
6057	 * means we reserve 2 more, this pushes us to allocate from the next
6058	 * larger slab size.
6059	 * i.e. RXBUFFER_2048 --> size-4096 slab
6060	 * However with the new *_jumbo_rx* routines, jumbo receives will use
6061	 * fragmented skbs
6062	 */
6063
6064	if (max_frame <= 2048)
6065		adapter->rx_buffer_len = 2048;
6066	else
6067		adapter->rx_buffer_len = 4096;
6068
6069	/* adjust allocation if LPE protects us, and we aren't using SBP */
6070	if (max_frame <= (VLAN_ETH_FRAME_LEN + ETH_FCS_LEN))
6071		adapter->rx_buffer_len = VLAN_ETH_FRAME_LEN + ETH_FCS_LEN;
6072
6073	if (netif_running(netdev))
6074		e1000e_up(adapter);
6075	else
6076		e1000e_reset(adapter);
6077
6078	pm_runtime_put_sync(netdev->dev.parent);
6079
6080	clear_bit(__E1000_RESETTING, &adapter->state);
6081
6082	return 0;
6083}
6084
6085static int e1000_mii_ioctl(struct net_device *netdev, struct ifreq *ifr,
6086			   int cmd)
6087{
6088	struct e1000_adapter *adapter = netdev_priv(netdev);
6089	struct mii_ioctl_data *data = if_mii(ifr);
6090
6091	if (adapter->hw.phy.media_type != e1000_media_type_copper)
6092		return -EOPNOTSUPP;
6093
6094	switch (cmd) {
6095	case SIOCGMIIPHY:
6096		data->phy_id = adapter->hw.phy.addr;
6097		break;
6098	case SIOCGMIIREG:
6099		e1000_phy_read_status(adapter);
6100
6101		switch (data->reg_num & 0x1F) {
6102		case MII_BMCR:
6103			data->val_out = adapter->phy_regs.bmcr;
6104			break;
6105		case MII_BMSR:
6106			data->val_out = adapter->phy_regs.bmsr;
6107			break;
6108		case MII_PHYSID1:
6109			data->val_out = (adapter->hw.phy.id >> 16);
6110			break;
6111		case MII_PHYSID2:
6112			data->val_out = (adapter->hw.phy.id & 0xFFFF);
6113			break;
6114		case MII_ADVERTISE:
6115			data->val_out = adapter->phy_regs.advertise;
6116			break;
6117		case MII_LPA:
6118			data->val_out = adapter->phy_regs.lpa;
6119			break;
6120		case MII_EXPANSION:
6121			data->val_out = adapter->phy_regs.expansion;
6122			break;
6123		case MII_CTRL1000:
6124			data->val_out = adapter->phy_regs.ctrl1000;
6125			break;
6126		case MII_STAT1000:
6127			data->val_out = adapter->phy_regs.stat1000;
6128			break;
6129		case MII_ESTATUS:
6130			data->val_out = adapter->phy_regs.estatus;
6131			break;
6132		default:
6133			return -EIO;
6134		}
6135		break;
6136	case SIOCSMIIREG:
6137	default:
6138		return -EOPNOTSUPP;
6139	}
6140	return 0;
6141}
6142
6143/**
6144 * e1000e_hwtstamp_ioctl - control hardware time stamping
6145 * @netdev: network interface device structure
6146 * @ifr: interface request
6147 *
6148 * Outgoing time stamping can be enabled and disabled. Play nice and
6149 * disable it when requested, although it shouldn't cause any overhead
6150 * when no packet needs it. At most one packet in the queue may be
6151 * marked for time stamping, otherwise it would be impossible to tell
6152 * for sure to which packet the hardware time stamp belongs.
6153 *
6154 * Incoming time stamping has to be configured via the hardware filters.
6155 * Not all combinations are supported, in particular event type has to be
6156 * specified. Matching the kind of event packet is not supported, with the
6157 * exception of "all V2 events regardless of level 2 or 4".
6158 **/
6159static int e1000e_hwtstamp_set(struct net_device *netdev, struct ifreq *ifr)
6160{
6161	struct e1000_adapter *adapter = netdev_priv(netdev);
6162	struct hwtstamp_config config;
6163	int ret_val;
6164
6165	if (copy_from_user(&config, ifr->ifr_data, sizeof(config)))
6166		return -EFAULT;
6167
6168	ret_val = e1000e_config_hwtstamp(adapter, &config);
6169	if (ret_val)
6170		return ret_val;
6171
6172	switch (config.rx_filter) {
6173	case HWTSTAMP_FILTER_PTP_V2_L4_SYNC:
6174	case HWTSTAMP_FILTER_PTP_V2_L2_SYNC:
6175	case HWTSTAMP_FILTER_PTP_V2_SYNC:
6176	case HWTSTAMP_FILTER_PTP_V2_L4_DELAY_REQ:
6177	case HWTSTAMP_FILTER_PTP_V2_L2_DELAY_REQ:
6178	case HWTSTAMP_FILTER_PTP_V2_DELAY_REQ:
6179		/* With V2 type filters which specify a Sync or Delay Request,
6180		 * Path Delay Request/Response messages are also time stamped
6181		 * by hardware so notify the caller the requested packets plus
6182		 * some others are time stamped.
6183		 */
6184		config.rx_filter = HWTSTAMP_FILTER_SOME;
6185		break;
6186	default:
6187		break;
6188	}
6189
6190	return copy_to_user(ifr->ifr_data, &config,
6191			    sizeof(config)) ? -EFAULT : 0;
6192}
6193
6194static int e1000e_hwtstamp_get(struct net_device *netdev, struct ifreq *ifr)
6195{
6196	struct e1000_adapter *adapter = netdev_priv(netdev);
6197
6198	return copy_to_user(ifr->ifr_data, &adapter->hwtstamp_config,
6199			    sizeof(adapter->hwtstamp_config)) ? -EFAULT : 0;
6200}
6201
6202static int e1000_ioctl(struct net_device *netdev, struct ifreq *ifr, int cmd)
6203{
6204	switch (cmd) {
6205	case SIOCGMIIPHY:
6206	case SIOCGMIIREG:
6207	case SIOCSMIIREG:
6208		return e1000_mii_ioctl(netdev, ifr, cmd);
6209	case SIOCSHWTSTAMP:
6210		return e1000e_hwtstamp_set(netdev, ifr);
6211	case SIOCGHWTSTAMP:
6212		return e1000e_hwtstamp_get(netdev, ifr);
6213	default:
6214		return -EOPNOTSUPP;
6215	}
6216}
6217
6218static int e1000_init_phy_wakeup(struct e1000_adapter *adapter, u32 wufc)
6219{
6220	struct e1000_hw *hw = &adapter->hw;
6221	u32 i, mac_reg, wuc;
6222	u16 phy_reg, wuc_enable;
6223	int retval;
6224
6225	/* copy MAC RARs to PHY RARs */
6226	e1000_copy_rx_addrs_to_phy_ich8lan(hw);
6227
6228	retval = hw->phy.ops.acquire(hw);
6229	if (retval) {
6230		e_err("Could not acquire PHY\n");
6231		return retval;
6232	}
6233
6234	/* Enable access to wakeup registers on and set page to BM_WUC_PAGE */
6235	retval = e1000_enable_phy_wakeup_reg_access_bm(hw, &wuc_enable);
6236	if (retval)
6237		goto release;
6238
6239	/* copy MAC MTA to PHY MTA - only needed for pchlan */
6240	for (i = 0; i < adapter->hw.mac.mta_reg_count; i++) {
6241		mac_reg = E1000_READ_REG_ARRAY(hw, E1000_MTA, i);
6242		hw->phy.ops.write_reg_page(hw, BM_MTA(i),
6243					   (u16)(mac_reg & 0xFFFF));
6244		hw->phy.ops.write_reg_page(hw, BM_MTA(i) + 1,
6245					   (u16)((mac_reg >> 16) & 0xFFFF));
6246	}
6247
6248	/* configure PHY Rx Control register */
6249	hw->phy.ops.read_reg_page(&adapter->hw, BM_RCTL, &phy_reg);
6250	mac_reg = er32(RCTL);
6251	if (mac_reg & E1000_RCTL_UPE)
6252		phy_reg |= BM_RCTL_UPE;
6253	if (mac_reg & E1000_RCTL_MPE)
6254		phy_reg |= BM_RCTL_MPE;
6255	phy_reg &= ~(BM_RCTL_MO_MASK);
6256	if (mac_reg & E1000_RCTL_MO_3)
6257		phy_reg |= (((mac_reg & E1000_RCTL_MO_3) >> E1000_RCTL_MO_SHIFT)
6258			    << BM_RCTL_MO_SHIFT);
6259	if (mac_reg & E1000_RCTL_BAM)
6260		phy_reg |= BM_RCTL_BAM;
6261	if (mac_reg & E1000_RCTL_PMCF)
6262		phy_reg |= BM_RCTL_PMCF;
6263	mac_reg = er32(CTRL);
6264	if (mac_reg & E1000_CTRL_RFCE)
6265		phy_reg |= BM_RCTL_RFCE;
6266	hw->phy.ops.write_reg_page(&adapter->hw, BM_RCTL, phy_reg);
6267
6268	wuc = E1000_WUC_PME_EN;
6269	if (wufc & (E1000_WUFC_MAG | E1000_WUFC_LNKC))
6270		wuc |= E1000_WUC_APME;
6271
6272	/* enable PHY wakeup in MAC register */
6273	ew32(WUFC, wufc);
6274	ew32(WUC, (E1000_WUC_PHY_WAKE | E1000_WUC_APMPME |
6275		   E1000_WUC_PME_STATUS | wuc));
6276
6277	/* configure and enable PHY wakeup in PHY registers */
6278	hw->phy.ops.write_reg_page(&adapter->hw, BM_WUFC, wufc);
6279	hw->phy.ops.write_reg_page(&adapter->hw, BM_WUC, wuc);
6280
6281	/* activate PHY wakeup */
6282	wuc_enable |= BM_WUC_ENABLE_BIT | BM_WUC_HOST_WU_BIT;
6283	retval = e1000_disable_phy_wakeup_reg_access_bm(hw, &wuc_enable);
6284	if (retval)
6285		e_err("Could not set PHY Host Wakeup bit\n");
6286release:
6287	hw->phy.ops.release(hw);
6288
6289	return retval;
6290}
6291
6292static void e1000e_flush_lpic(struct pci_dev *pdev)
6293{
6294	struct net_device *netdev = pci_get_drvdata(pdev);
6295	struct e1000_adapter *adapter = netdev_priv(netdev);
6296	struct e1000_hw *hw = &adapter->hw;
6297	u32 ret_val;
6298
6299	pm_runtime_get_sync(netdev->dev.parent);
6300
6301	ret_val = hw->phy.ops.acquire(hw);
6302	if (ret_val)
6303		goto fl_out;
6304
6305	pr_info("EEE TX LPI TIMER: %08X\n",
6306		er32(LPIC) >> E1000_LPIC_LPIET_SHIFT);
6307
6308	hw->phy.ops.release(hw);
6309
6310fl_out:
6311	pm_runtime_put_sync(netdev->dev.parent);
6312}
6313
6314/* S0ix implementation */
6315static void e1000e_s0ix_entry_flow(struct e1000_adapter *adapter)
6316{
6317	struct e1000_hw *hw = &adapter->hw;
6318	u32 mac_data;
6319	u16 phy_data;
6320
6321	/* Disable the periodic inband message,
6322	 * don't request PCIe clock in K1 page770_17[10:9] = 10b
6323	 */
6324	e1e_rphy(hw, HV_PM_CTRL, &phy_data);
6325	phy_data &= ~HV_PM_CTRL_K1_CLK_REQ;
6326	phy_data |= BIT(10);
6327	e1e_wphy(hw, HV_PM_CTRL, phy_data);
6328
6329	/* Make sure we don't exit K1 every time a new packet arrives
6330	 * 772_29[5] = 1 CS_Mode_Stay_In_K1
6331	 */
6332	e1e_rphy(hw, I217_CGFREG, &phy_data);
6333	phy_data |= BIT(5);
6334	e1e_wphy(hw, I217_CGFREG, phy_data);
6335
6336	/* Change the MAC/PHY interface to SMBus
6337	 * Force the SMBus in PHY page769_23[0] = 1
6338	 * Force the SMBus in MAC CTRL_EXT[11] = 1
6339	 */
6340	e1e_rphy(hw, CV_SMB_CTRL, &phy_data);
6341	phy_data |= CV_SMB_CTRL_FORCE_SMBUS;
6342	e1e_wphy(hw, CV_SMB_CTRL, phy_data);
6343	mac_data = er32(CTRL_EXT);
6344	mac_data |= E1000_CTRL_EXT_FORCE_SMBUS;
6345	ew32(CTRL_EXT, mac_data);
6346
6347	/* DFT control: PHY bit: page769_20[0] = 1
6348	 * Gate PPW via EXTCNF_CTRL - set 0x0F00[7] = 1
6349	 */
6350	e1e_rphy(hw, I82579_DFT_CTRL, &phy_data);
6351	phy_data |= BIT(0);
6352	e1e_wphy(hw, I82579_DFT_CTRL, phy_data);
6353
6354	mac_data = er32(EXTCNF_CTRL);
6355	mac_data |= E1000_EXTCNF_CTRL_GATE_PHY_CFG;
6356	ew32(EXTCNF_CTRL, mac_data);
6357
6358	/* Check MAC Tx/Rx packet buffer pointers.
6359	 * Reset MAC Tx/Rx packet buffer pointers to suppress any
6360	 * pending traffic indication that would prevent power gating.
6361	 */
6362	mac_data = er32(TDFH);
6363	if (mac_data)
6364		ew32(TDFH, 0);
6365	mac_data = er32(TDFT);
6366	if (mac_data)
6367		ew32(TDFT, 0);
6368	mac_data = er32(TDFHS);
6369	if (mac_data)
6370		ew32(TDFHS, 0);
6371	mac_data = er32(TDFTS);
6372	if (mac_data)
6373		ew32(TDFTS, 0);
6374	mac_data = er32(TDFPC);
6375	if (mac_data)
6376		ew32(TDFPC, 0);
6377	mac_data = er32(RDFH);
6378	if (mac_data)
6379		ew32(RDFH, 0);
6380	mac_data = er32(RDFT);
6381	if (mac_data)
6382		ew32(RDFT, 0);
6383	mac_data = er32(RDFHS);
6384	if (mac_data)
6385		ew32(RDFHS, 0);
6386	mac_data = er32(RDFTS);
6387	if (mac_data)
6388		ew32(RDFTS, 0);
6389	mac_data = er32(RDFPC);
6390	if (mac_data)
6391		ew32(RDFPC, 0);
6392
6393	/* Enable the Dynamic Power Gating in the MAC */
6394	mac_data = er32(FEXTNVM7);
6395	mac_data |= BIT(22);
6396	ew32(FEXTNVM7, mac_data);
6397
6398	/* Disable the time synchronization clock */
6399	mac_data = er32(FEXTNVM7);
6400	mac_data |= BIT(31);
6401	mac_data &= ~BIT(0);
6402	ew32(FEXTNVM7, mac_data);
6403
6404	/* Dynamic Power Gating Enable */
6405	mac_data = er32(CTRL_EXT);
6406	mac_data |= BIT(3);
6407	ew32(CTRL_EXT, mac_data);
6408
6409	/* Disable disconnected cable conditioning for Power Gating */
6410	mac_data = er32(DPGFR);
6411	mac_data |= BIT(2);
6412	ew32(DPGFR, mac_data);
6413
6414	/* Don't wake from dynamic Power Gating with clock request */
6415	mac_data = er32(FEXTNVM12);
6416	mac_data |= BIT(12);
6417	ew32(FEXTNVM12, mac_data);
6418
6419	/* Ungate PGCB clock */
6420	mac_data = er32(FEXTNVM9);
6421	mac_data &= ~BIT(28);
6422	ew32(FEXTNVM9, mac_data);
6423
6424	/* Enable K1 off to enable mPHY Power Gating */
6425	mac_data = er32(FEXTNVM6);
6426	mac_data |= BIT(31);
6427	ew32(FEXTNVM6, mac_data);
6428
6429	/* Enable mPHY power gating for any link and speed */
6430	mac_data = er32(FEXTNVM8);
6431	mac_data |= BIT(9);
6432	ew32(FEXTNVM8, mac_data);
6433
6434	/* Enable the Dynamic Clock Gating in the DMA and MAC */
6435	mac_data = er32(CTRL_EXT);
6436	mac_data |= E1000_CTRL_EXT_DMA_DYN_CLK_EN;
6437	ew32(CTRL_EXT, mac_data);
6438
6439	/* No MAC DPG gating SLP_S0 in modern standby
6440	 * Switch the logic of the lanphypc to use PMC counter
6441	 */
6442	mac_data = er32(FEXTNVM5);
6443	mac_data |= BIT(7);
6444	ew32(FEXTNVM5, mac_data);
6445}
6446
6447static void e1000e_s0ix_exit_flow(struct e1000_adapter *adapter)
6448{
6449	struct e1000_hw *hw = &adapter->hw;
6450	u32 mac_data;
6451	u16 phy_data;
6452
6453	/* Disable the Dynamic Power Gating in the MAC */
6454	mac_data = er32(FEXTNVM7);
6455	mac_data &= 0xFFBFFFFF;
6456	ew32(FEXTNVM7, mac_data);
6457
6458	/* Enable the time synchronization clock */
6459	mac_data = er32(FEXTNVM7);
6460	mac_data |= BIT(0);
6461	ew32(FEXTNVM7, mac_data);
6462
6463	/* Disable mPHY power gating for any link and speed */
6464	mac_data = er32(FEXTNVM8);
6465	mac_data &= ~BIT(9);
6466	ew32(FEXTNVM8, mac_data);
6467
6468	/* Disable K1 off */
6469	mac_data = er32(FEXTNVM6);
6470	mac_data &= ~BIT(31);
6471	ew32(FEXTNVM6, mac_data);
6472
6473	/* Disable Ungate PGCB clock */
6474	mac_data = er32(FEXTNVM9);
6475	mac_data |= BIT(28);
6476	ew32(FEXTNVM9, mac_data);
6477
6478	/* Cancel not waking from dynamic
6479	 * Power Gating with clock request
6480	 */
6481	mac_data = er32(FEXTNVM12);
6482	mac_data &= ~BIT(12);
6483	ew32(FEXTNVM12, mac_data);
6484
6485	/* Cancel disable disconnected cable conditioning
6486	 * for Power Gating
6487	 */
6488	mac_data = er32(DPGFR);
6489	mac_data &= ~BIT(2);
6490	ew32(DPGFR, mac_data);
6491
6492	/* Disable Dynamic Power Gating */
6493	mac_data = er32(CTRL_EXT);
6494	mac_data &= 0xFFFFFFF7;
6495	ew32(CTRL_EXT, mac_data);
6496
6497	/* Disable the Dynamic Clock Gating in the DMA and MAC */
6498	mac_data = er32(CTRL_EXT);
6499	mac_data &= 0xFFF7FFFF;
6500	ew32(CTRL_EXT, mac_data);
6501
6502	/* Revert the lanphypc logic to use the internal Gbe counter
6503	 * and not the PMC counter
6504	 */
6505	mac_data = er32(FEXTNVM5);
6506	mac_data &= 0xFFFFFF7F;
6507	ew32(FEXTNVM5, mac_data);
6508
6509	/* Enable the periodic inband message,
6510	 * Request PCIe clock in K1 page770_17[10:9] =01b
6511	 */
6512	e1e_rphy(hw, HV_PM_CTRL, &phy_data);
6513	phy_data &= 0xFBFF;
6514	phy_data |= HV_PM_CTRL_K1_CLK_REQ;
6515	e1e_wphy(hw, HV_PM_CTRL, phy_data);
6516
6517	/* Return back configuration
6518	 * 772_29[5] = 0 CS_Mode_Stay_In_K1
6519	 */
6520	e1e_rphy(hw, I217_CGFREG, &phy_data);
6521	phy_data &= 0xFFDF;
6522	e1e_wphy(hw, I217_CGFREG, phy_data);
6523
6524	/* Change the MAC/PHY interface to Kumeran
6525	 * Unforce the SMBus in PHY page769_23[0] = 0
6526	 * Unforce the SMBus in MAC CTRL_EXT[11] = 0
6527	 */
6528	e1e_rphy(hw, CV_SMB_CTRL, &phy_data);
6529	phy_data &= ~CV_SMB_CTRL_FORCE_SMBUS;
6530	e1e_wphy(hw, CV_SMB_CTRL, phy_data);
6531	mac_data = er32(CTRL_EXT);
6532	mac_data &= ~E1000_CTRL_EXT_FORCE_SMBUS;
6533	ew32(CTRL_EXT, mac_data);
6534}
6535
6536static int e1000e_pm_freeze(struct device *dev)
6537{
6538	struct net_device *netdev = dev_get_drvdata(dev);
6539	struct e1000_adapter *adapter = netdev_priv(netdev);
6540	bool present;
6541
6542	rtnl_lock();
6543
6544	present = netif_device_present(netdev);
6545	netif_device_detach(netdev);
6546
6547	if (present && netif_running(netdev)) {
6548		int count = E1000_CHECK_RESET_COUNT;
6549
6550		while (test_bit(__E1000_RESETTING, &adapter->state) && count--)
6551			usleep_range(10000, 11000);
6552
6553		WARN_ON(test_bit(__E1000_RESETTING, &adapter->state));
6554
6555		/* Quiesce the device without resetting the hardware */
6556		e1000e_down(adapter, false);
6557		e1000_free_irq(adapter);
6558	}
6559	rtnl_unlock();
6560
6561	e1000e_reset_interrupt_capability(adapter);
6562
6563	/* Allow time for pending master requests to run */
6564	e1000e_disable_pcie_master(&adapter->hw);
6565
6566	return 0;
6567}
6568
6569static int __e1000_shutdown(struct pci_dev *pdev, bool runtime)
6570{
6571	struct net_device *netdev = pci_get_drvdata(pdev);
6572	struct e1000_adapter *adapter = netdev_priv(netdev);
6573	struct e1000_hw *hw = &adapter->hw;
6574	u32 ctrl, ctrl_ext, rctl, status, wufc;
6575	int retval = 0;
6576
6577	/* Runtime suspend should only enable wakeup for link changes */
6578	if (runtime)
6579		wufc = E1000_WUFC_LNKC;
6580	else if (device_may_wakeup(&pdev->dev))
6581		wufc = adapter->wol;
6582	else
6583		wufc = 0;
6584
6585	status = er32(STATUS);
6586	if (status & E1000_STATUS_LU)
6587		wufc &= ~E1000_WUFC_LNKC;
6588
6589	if (wufc) {
6590		e1000_setup_rctl(adapter);
6591		e1000e_set_rx_mode(netdev);
6592
6593		/* turn on all-multi mode if wake on multicast is enabled */
6594		if (wufc & E1000_WUFC_MC) {
6595			rctl = er32(RCTL);
6596			rctl |= E1000_RCTL_MPE;
6597			ew32(RCTL, rctl);
6598		}
6599
6600		ctrl = er32(CTRL);
6601		ctrl |= E1000_CTRL_ADVD3WUC;
6602		if (!(adapter->flags2 & FLAG2_HAS_PHY_WAKEUP))
6603			ctrl |= E1000_CTRL_EN_PHY_PWR_MGMT;
6604		ew32(CTRL, ctrl);
6605
6606		if (adapter->hw.phy.media_type == e1000_media_type_fiber ||
6607		    adapter->hw.phy.media_type ==
6608		    e1000_media_type_internal_serdes) {
6609			/* keep the laser running in D3 */
6610			ctrl_ext = er32(CTRL_EXT);
6611			ctrl_ext |= E1000_CTRL_EXT_SDP3_DATA;
6612			ew32(CTRL_EXT, ctrl_ext);
6613		}
6614
6615		if (!runtime)
6616			e1000e_power_up_phy(adapter);
6617
6618		if (adapter->flags & FLAG_IS_ICH)
6619			e1000_suspend_workarounds_ich8lan(&adapter->hw);
6620
6621		if (adapter->flags2 & FLAG2_HAS_PHY_WAKEUP) {
6622			/* enable wakeup by the PHY */
6623			retval = e1000_init_phy_wakeup(adapter, wufc);
6624			if (retval)
6625				return retval;
6626		} else {
6627			/* enable wakeup by the MAC */
6628			ew32(WUFC, wufc);
6629			ew32(WUC, E1000_WUC_PME_EN);
6630		}
6631	} else {
6632		ew32(WUC, 0);
6633		ew32(WUFC, 0);
6634
6635		e1000_power_down_phy(adapter);
6636	}
6637
6638	if (adapter->hw.phy.type == e1000_phy_igp_3) {
6639		e1000e_igp3_phy_powerdown_workaround_ich8lan(&adapter->hw);
6640	} else if (hw->mac.type >= e1000_pch_lpt) {
6641		if (wufc && !(wufc & (E1000_WUFC_EX | E1000_WUFC_MC | E1000_WUFC_BC)))
6642			/* ULP does not support wake from unicast, multicast
6643			 * or broadcast.
6644			 */
6645			retval = e1000_enable_ulp_lpt_lp(hw, !runtime);
6646
6647		if (retval)
6648			return retval;
6649	}
6650
6651	/* Ensure that the appropriate bits are set in LPI_CTRL
6652	 * for EEE in Sx
6653	 */
6654	if ((hw->phy.type >= e1000_phy_i217) &&
6655	    adapter->eee_advert && hw->dev_spec.ich8lan.eee_lp_ability) {
6656		u16 lpi_ctrl = 0;
6657
6658		retval = hw->phy.ops.acquire(hw);
6659		if (!retval) {
6660			retval = e1e_rphy_locked(hw, I82579_LPI_CTRL,
6661						 &lpi_ctrl);
6662			if (!retval) {
6663				if (adapter->eee_advert &
6664				    hw->dev_spec.ich8lan.eee_lp_ability &
6665				    I82579_EEE_100_SUPPORTED)
6666					lpi_ctrl |= I82579_LPI_CTRL_100_ENABLE;
6667				if (adapter->eee_advert &
6668				    hw->dev_spec.ich8lan.eee_lp_ability &
6669				    I82579_EEE_1000_SUPPORTED)
6670					lpi_ctrl |= I82579_LPI_CTRL_1000_ENABLE;
6671
6672				retval = e1e_wphy_locked(hw, I82579_LPI_CTRL,
6673							 lpi_ctrl);
6674			}
6675		}
6676		hw->phy.ops.release(hw);
6677	}
6678
6679	/* Release control of h/w to f/w.  If f/w is AMT enabled, this
6680	 * would have already happened in close and is redundant.
6681	 */
6682	e1000e_release_hw_control(adapter);
6683
6684	pci_clear_master(pdev);
6685
6686	/* The pci-e switch on some quad port adapters will report a
6687	 * correctable error when the MAC transitions from D0 to D3.  To
6688	 * prevent this we need to mask off the correctable errors on the
6689	 * downstream port of the pci-e switch.
6690	 *
6691	 * We don't have the associated upstream bridge while assigning
6692	 * the PCI device into guest. For example, the KVM on power is
6693	 * one of the cases.
6694	 */
6695	if (adapter->flags & FLAG_IS_QUAD_PORT) {
6696		struct pci_dev *us_dev = pdev->bus->self;
6697		u16 devctl;
6698
6699		if (!us_dev)
6700			return 0;
6701
6702		pcie_capability_read_word(us_dev, PCI_EXP_DEVCTL, &devctl);
6703		pcie_capability_write_word(us_dev, PCI_EXP_DEVCTL,
6704					   (devctl & ~PCI_EXP_DEVCTL_CERE));
6705
6706		pci_save_state(pdev);
6707		pci_prepare_to_sleep(pdev);
6708
6709		pcie_capability_write_word(us_dev, PCI_EXP_DEVCTL, devctl);
6710	}
6711
6712	return 0;
6713}
6714
6715/**
6716 * __e1000e_disable_aspm - Disable ASPM states
6717 * @pdev: pointer to PCI device struct
6718 * @state: bit-mask of ASPM states to disable
6719 * @locked: indication if this context holds pci_bus_sem locked.
6720 *
6721 * Some devices *must* have certain ASPM states disabled per hardware errata.
6722 **/
6723static void __e1000e_disable_aspm(struct pci_dev *pdev, u16 state, int locked)
6724{
6725	struct pci_dev *parent = pdev->bus->self;
6726	u16 aspm_dis_mask = 0;
6727	u16 pdev_aspmc, parent_aspmc;
6728
6729	switch (state) {
6730	case PCIE_LINK_STATE_L0S:
6731	case PCIE_LINK_STATE_L0S | PCIE_LINK_STATE_L1:
6732		aspm_dis_mask |= PCI_EXP_LNKCTL_ASPM_L0S;
6733		fallthrough; /* can't have L1 without L0s */
6734	case PCIE_LINK_STATE_L1:
6735		aspm_dis_mask |= PCI_EXP_LNKCTL_ASPM_L1;
6736		break;
6737	default:
6738		return;
6739	}
6740
6741	pcie_capability_read_word(pdev, PCI_EXP_LNKCTL, &pdev_aspmc);
6742	pdev_aspmc &= PCI_EXP_LNKCTL_ASPMC;
6743
6744	if (parent) {
6745		pcie_capability_read_word(parent, PCI_EXP_LNKCTL,
6746					  &parent_aspmc);
6747		parent_aspmc &= PCI_EXP_LNKCTL_ASPMC;
6748	}
6749
6750	/* Nothing to do if the ASPM states to be disabled already are */
6751	if (!(pdev_aspmc & aspm_dis_mask) &&
6752	    (!parent || !(parent_aspmc & aspm_dis_mask)))
6753		return;
6754
6755	dev_info(&pdev->dev, "Disabling ASPM %s %s\n",
6756		 (aspm_dis_mask & pdev_aspmc & PCI_EXP_LNKCTL_ASPM_L0S) ?
6757		 "L0s" : "",
6758		 (aspm_dis_mask & pdev_aspmc & PCI_EXP_LNKCTL_ASPM_L1) ?
6759		 "L1" : "");
6760
6761#ifdef CONFIG_PCIEASPM
6762	if (locked)
6763		pci_disable_link_state_locked(pdev, state);
6764	else
6765		pci_disable_link_state(pdev, state);
6766
6767	/* Double-check ASPM control.  If not disabled by the above, the
6768	 * BIOS is preventing that from happening (or CONFIG_PCIEASPM is
6769	 * not enabled); override by writing PCI config space directly.
6770	 */
6771	pcie_capability_read_word(pdev, PCI_EXP_LNKCTL, &pdev_aspmc);
6772	pdev_aspmc &= PCI_EXP_LNKCTL_ASPMC;
6773
6774	if (!(aspm_dis_mask & pdev_aspmc))
6775		return;
6776#endif
6777
6778	/* Both device and parent should have the same ASPM setting.
6779	 * Disable ASPM in downstream component first and then upstream.
6780	 */
6781	pcie_capability_clear_word(pdev, PCI_EXP_LNKCTL, aspm_dis_mask);
6782
6783	if (parent)
6784		pcie_capability_clear_word(parent, PCI_EXP_LNKCTL,
6785					   aspm_dis_mask);
6786}
6787
6788/**
6789 * e1000e_disable_aspm - Disable ASPM states.
6790 * @pdev: pointer to PCI device struct
6791 * @state: bit-mask of ASPM states to disable
6792 *
6793 * This function acquires the pci_bus_sem!
6794 * Some devices *must* have certain ASPM states disabled per hardware errata.
6795 **/
6796static void e1000e_disable_aspm(struct pci_dev *pdev, u16 state)
6797{
6798	__e1000e_disable_aspm(pdev, state, 0);
6799}
6800
6801/**
6802 * e1000e_disable_aspm_locked   Disable ASPM states.
6803 * @pdev: pointer to PCI device struct
6804 * @state: bit-mask of ASPM states to disable
6805 *
6806 * This function must be called with pci_bus_sem acquired!
6807 * Some devices *must* have certain ASPM states disabled per hardware errata.
6808 **/
6809static void e1000e_disable_aspm_locked(struct pci_dev *pdev, u16 state)
6810{
6811	__e1000e_disable_aspm(pdev, state, 1);
6812}
6813
6814static int e1000e_pm_thaw(struct device *dev)
6815{
6816	struct net_device *netdev = dev_get_drvdata(dev);
6817	struct e1000_adapter *adapter = netdev_priv(netdev);
6818	int rc = 0;
6819
6820	e1000e_set_interrupt_capability(adapter);
6821
6822	rtnl_lock();
6823	if (netif_running(netdev)) {
6824		rc = e1000_request_irq(adapter);
6825		if (rc)
6826			goto err_irq;
6827
6828		e1000e_up(adapter);
6829	}
6830
6831	netif_device_attach(netdev);
6832err_irq:
6833	rtnl_unlock();
6834
6835	return rc;
6836}
6837
6838static int __e1000_resume(struct pci_dev *pdev)
6839{
6840	struct net_device *netdev = pci_get_drvdata(pdev);
6841	struct e1000_adapter *adapter = netdev_priv(netdev);
6842	struct e1000_hw *hw = &adapter->hw;
6843	u16 aspm_disable_flag = 0;
6844
6845	if (adapter->flags2 & FLAG2_DISABLE_ASPM_L0S)
6846		aspm_disable_flag = PCIE_LINK_STATE_L0S;
6847	if (adapter->flags2 & FLAG2_DISABLE_ASPM_L1)
6848		aspm_disable_flag |= PCIE_LINK_STATE_L1;
6849	if (aspm_disable_flag)
6850		e1000e_disable_aspm(pdev, aspm_disable_flag);
6851
6852	pci_set_master(pdev);
6853
6854	if (hw->mac.type >= e1000_pch2lan)
6855		e1000_resume_workarounds_pchlan(&adapter->hw);
6856
6857	e1000e_power_up_phy(adapter);
6858
6859	/* report the system wakeup cause from S3/S4 */
6860	if (adapter->flags2 & FLAG2_HAS_PHY_WAKEUP) {
6861		u16 phy_data;
6862
6863		e1e_rphy(&adapter->hw, BM_WUS, &phy_data);
6864		if (phy_data) {
6865			e_info("PHY Wakeup cause - %s\n",
6866			       phy_data & E1000_WUS_EX ? "Unicast Packet" :
6867			       phy_data & E1000_WUS_MC ? "Multicast Packet" :
6868			       phy_data & E1000_WUS_BC ? "Broadcast Packet" :
6869			       phy_data & E1000_WUS_MAG ? "Magic Packet" :
6870			       phy_data & E1000_WUS_LNKC ?
6871			       "Link Status Change" : "other");
6872		}
6873		e1e_wphy(&adapter->hw, BM_WUS, ~0);
6874	} else {
6875		u32 wus = er32(WUS);
6876
6877		if (wus) {
6878			e_info("MAC Wakeup cause - %s\n",
6879			       wus & E1000_WUS_EX ? "Unicast Packet" :
6880			       wus & E1000_WUS_MC ? "Multicast Packet" :
6881			       wus & E1000_WUS_BC ? "Broadcast Packet" :
6882			       wus & E1000_WUS_MAG ? "Magic Packet" :
6883			       wus & E1000_WUS_LNKC ? "Link Status Change" :
6884			       "other");
6885		}
6886		ew32(WUS, ~0);
6887	}
6888
6889	e1000e_reset(adapter);
6890
6891	e1000_init_manageability_pt(adapter);
6892
6893	/* If the controller has AMT, do not set DRV_LOAD until the interface
6894	 * is up.  For all other cases, let the f/w know that the h/w is now
6895	 * under the control of the driver.
6896	 */
6897	if (!(adapter->flags & FLAG_HAS_AMT))
6898		e1000e_get_hw_control(adapter);
6899
6900	return 0;
6901}
6902
6903static __maybe_unused int e1000e_pm_suspend(struct device *dev)
6904{
6905	struct net_device *netdev = pci_get_drvdata(to_pci_dev(dev));
6906	struct e1000_adapter *adapter = netdev_priv(netdev);
6907	struct pci_dev *pdev = to_pci_dev(dev);
6908	int rc;
6909
6910	e1000e_flush_lpic(pdev);
6911
6912	e1000e_pm_freeze(dev);
6913
6914	rc = __e1000_shutdown(pdev, false);
6915	if (rc) {
6916		e1000e_pm_thaw(dev);
6917	} else {
6918		/* Introduce S0ix implementation */
6919		if (adapter->flags2 & FLAG2_ENABLE_S0IX_FLOWS)
6920			e1000e_s0ix_entry_flow(adapter);
6921	}
6922
6923	return rc;
6924}
6925
6926static __maybe_unused int e1000e_pm_resume(struct device *dev)
6927{
6928	struct net_device *netdev = pci_get_drvdata(to_pci_dev(dev));
6929	struct e1000_adapter *adapter = netdev_priv(netdev);
6930	struct pci_dev *pdev = to_pci_dev(dev);
6931	int rc;
6932
6933	/* Introduce S0ix implementation */
6934	if (adapter->flags2 & FLAG2_ENABLE_S0IX_FLOWS)
6935		e1000e_s0ix_exit_flow(adapter);
6936
6937	rc = __e1000_resume(pdev);
6938	if (rc)
6939		return rc;
6940
6941	return e1000e_pm_thaw(dev);
6942}
6943
6944static __maybe_unused int e1000e_pm_runtime_idle(struct device *dev)
6945{
6946	struct net_device *netdev = dev_get_drvdata(dev);
6947	struct e1000_adapter *adapter = netdev_priv(netdev);
6948	u16 eee_lp;
6949
6950	eee_lp = adapter->hw.dev_spec.ich8lan.eee_lp_ability;
6951
6952	if (!e1000e_has_link(adapter)) {
6953		adapter->hw.dev_spec.ich8lan.eee_lp_ability = eee_lp;
6954		pm_schedule_suspend(dev, 5 * MSEC_PER_SEC);
6955	}
6956
6957	return -EBUSY;
6958}
6959
6960static __maybe_unused int e1000e_pm_runtime_resume(struct device *dev)
6961{
6962	struct pci_dev *pdev = to_pci_dev(dev);
6963	struct net_device *netdev = pci_get_drvdata(pdev);
6964	struct e1000_adapter *adapter = netdev_priv(netdev);
6965	int rc;
6966
6967	rc = __e1000_resume(pdev);
6968	if (rc)
6969		return rc;
6970
6971	if (netdev->flags & IFF_UP)
6972		e1000e_up(adapter);
6973
6974	return rc;
6975}
6976
6977static __maybe_unused int e1000e_pm_runtime_suspend(struct device *dev)
6978{
6979	struct pci_dev *pdev = to_pci_dev(dev);
6980	struct net_device *netdev = pci_get_drvdata(pdev);
6981	struct e1000_adapter *adapter = netdev_priv(netdev);
6982
6983	if (netdev->flags & IFF_UP) {
6984		int count = E1000_CHECK_RESET_COUNT;
6985
6986		while (test_bit(__E1000_RESETTING, &adapter->state) && count--)
6987			usleep_range(10000, 11000);
6988
6989		WARN_ON(test_bit(__E1000_RESETTING, &adapter->state));
6990
6991		/* Down the device without resetting the hardware */
6992		e1000e_down(adapter, false);
6993	}
6994
6995	if (__e1000_shutdown(pdev, true)) {
6996		e1000e_pm_runtime_resume(dev);
6997		return -EBUSY;
6998	}
6999
7000	return 0;
7001}
7002
7003static void e1000_shutdown(struct pci_dev *pdev)
7004{
7005	e1000e_flush_lpic(pdev);
7006
7007	e1000e_pm_freeze(&pdev->dev);
7008
7009	__e1000_shutdown(pdev, false);
7010}
7011
7012#ifdef CONFIG_NET_POLL_CONTROLLER
7013
7014static irqreturn_t e1000_intr_msix(int __always_unused irq, void *data)
7015{
7016	struct net_device *netdev = data;
7017	struct e1000_adapter *adapter = netdev_priv(netdev);
7018
7019	if (adapter->msix_entries) {
7020		int vector, msix_irq;
7021
7022		vector = 0;
7023		msix_irq = adapter->msix_entries[vector].vector;
7024		if (disable_hardirq(msix_irq))
7025			e1000_intr_msix_rx(msix_irq, netdev);
7026		enable_irq(msix_irq);
7027
7028		vector++;
7029		msix_irq = adapter->msix_entries[vector].vector;
7030		if (disable_hardirq(msix_irq))
7031			e1000_intr_msix_tx(msix_irq, netdev);
7032		enable_irq(msix_irq);
7033
7034		vector++;
7035		msix_irq = adapter->msix_entries[vector].vector;
7036		if (disable_hardirq(msix_irq))
7037			e1000_msix_other(msix_irq, netdev);
7038		enable_irq(msix_irq);
7039	}
7040
7041	return IRQ_HANDLED;
7042}
7043
7044/**
7045 * e1000_netpoll
7046 * @netdev: network interface device structure
7047 *
7048 * Polling 'interrupt' - used by things like netconsole to send skbs
7049 * without having to re-enable interrupts. It's not called while
7050 * the interrupt routine is executing.
7051 */
7052static void e1000_netpoll(struct net_device *netdev)
7053{
7054	struct e1000_adapter *adapter = netdev_priv(netdev);
7055
7056	switch (adapter->int_mode) {
7057	case E1000E_INT_MODE_MSIX:
7058		e1000_intr_msix(adapter->pdev->irq, netdev);
7059		break;
7060	case E1000E_INT_MODE_MSI:
7061		if (disable_hardirq(adapter->pdev->irq))
7062			e1000_intr_msi(adapter->pdev->irq, netdev);
7063		enable_irq(adapter->pdev->irq);
7064		break;
7065	default:		/* E1000E_INT_MODE_LEGACY */
7066		if (disable_hardirq(adapter->pdev->irq))
7067			e1000_intr(adapter->pdev->irq, netdev);
7068		enable_irq(adapter->pdev->irq);
7069		break;
7070	}
7071}
7072#endif
7073
7074/**
7075 * e1000_io_error_detected - called when PCI error is detected
7076 * @pdev: Pointer to PCI device
7077 * @state: The current pci connection state
7078 *
7079 * This function is called after a PCI bus error affecting
7080 * this device has been detected.
7081 */
7082static pci_ers_result_t e1000_io_error_detected(struct pci_dev *pdev,
7083						pci_channel_state_t state)
7084{
7085	e1000e_pm_freeze(&pdev->dev);
7086
7087	if (state == pci_channel_io_perm_failure)
7088		return PCI_ERS_RESULT_DISCONNECT;
7089
7090	pci_disable_device(pdev);
7091
7092	/* Request a slot slot reset. */
7093	return PCI_ERS_RESULT_NEED_RESET;
7094}
7095
7096/**
7097 * e1000_io_slot_reset - called after the pci bus has been reset.
7098 * @pdev: Pointer to PCI device
7099 *
7100 * Restart the card from scratch, as if from a cold-boot. Implementation
7101 * resembles the first-half of the e1000e_pm_resume routine.
7102 */
7103static pci_ers_result_t e1000_io_slot_reset(struct pci_dev *pdev)
7104{
7105	struct net_device *netdev = pci_get_drvdata(pdev);
7106	struct e1000_adapter *adapter = netdev_priv(netdev);
7107	struct e1000_hw *hw = &adapter->hw;
7108	u16 aspm_disable_flag = 0;
7109	int err;
7110	pci_ers_result_t result;
7111
7112	if (adapter->flags2 & FLAG2_DISABLE_ASPM_L0S)
7113		aspm_disable_flag = PCIE_LINK_STATE_L0S;
7114	if (adapter->flags2 & FLAG2_DISABLE_ASPM_L1)
7115		aspm_disable_flag |= PCIE_LINK_STATE_L1;
7116	if (aspm_disable_flag)
7117		e1000e_disable_aspm_locked(pdev, aspm_disable_flag);
7118
7119	err = pci_enable_device_mem(pdev);
7120	if (err) {
7121		dev_err(&pdev->dev,
7122			"Cannot re-enable PCI device after reset.\n");
7123		result = PCI_ERS_RESULT_DISCONNECT;
7124	} else {
7125		pdev->state_saved = true;
7126		pci_restore_state(pdev);
7127		pci_set_master(pdev);
7128
7129		pci_enable_wake(pdev, PCI_D3hot, 0);
7130		pci_enable_wake(pdev, PCI_D3cold, 0);
7131
7132		e1000e_reset(adapter);
7133		ew32(WUS, ~0);
7134		result = PCI_ERS_RESULT_RECOVERED;
7135	}
7136
7137	return result;
7138}
7139
7140/**
7141 * e1000_io_resume - called when traffic can start flowing again.
7142 * @pdev: Pointer to PCI device
7143 *
7144 * This callback is called when the error recovery driver tells us that
7145 * its OK to resume normal operation. Implementation resembles the
7146 * second-half of the e1000e_pm_resume routine.
7147 */
7148static void e1000_io_resume(struct pci_dev *pdev)
7149{
7150	struct net_device *netdev = pci_get_drvdata(pdev);
7151	struct e1000_adapter *adapter = netdev_priv(netdev);
7152
7153	e1000_init_manageability_pt(adapter);
7154
7155	e1000e_pm_thaw(&pdev->dev);
7156
7157	/* If the controller has AMT, do not set DRV_LOAD until the interface
7158	 * is up.  For all other cases, let the f/w know that the h/w is now
7159	 * under the control of the driver.
7160	 */
7161	if (!(adapter->flags & FLAG_HAS_AMT))
7162		e1000e_get_hw_control(adapter);
7163}
7164
7165static void e1000_print_device_info(struct e1000_adapter *adapter)
7166{
7167	struct e1000_hw *hw = &adapter->hw;
7168	struct net_device *netdev = adapter->netdev;
7169	u32 ret_val;
7170	u8 pba_str[E1000_PBANUM_LENGTH];
7171
7172	/* print bus type/speed/width info */
7173	e_info("(PCI Express:2.5GT/s:%s) %pM\n",
7174	       /* bus width */
7175	       ((hw->bus.width == e1000_bus_width_pcie_x4) ? "Width x4" :
7176		"Width x1"),
7177	       /* MAC address */
7178	       netdev->dev_addr);
7179	e_info("Intel(R) PRO/%s Network Connection\n",
7180	       (hw->phy.type == e1000_phy_ife) ? "10/100" : "1000");
7181	ret_val = e1000_read_pba_string_generic(hw, pba_str,
7182						E1000_PBANUM_LENGTH);
7183	if (ret_val)
7184		strlcpy((char *)pba_str, "Unknown", sizeof(pba_str));
7185	e_info("MAC: %d, PHY: %d, PBA No: %s\n",
7186	       hw->mac.type, hw->phy.type, pba_str);
7187}
7188
7189static void e1000_eeprom_checks(struct e1000_adapter *adapter)
7190{
7191	struct e1000_hw *hw = &adapter->hw;
7192	int ret_val;
7193	u16 buf = 0;
7194
7195	if (hw->mac.type != e1000_82573)
7196		return;
7197
7198	ret_val = e1000_read_nvm(hw, NVM_INIT_CONTROL2_REG, 1, &buf);
7199	le16_to_cpus(&buf);
7200	if (!ret_val && (!(buf & BIT(0)))) {
7201		/* Deep Smart Power Down (DSPD) */
7202		dev_warn(&adapter->pdev->dev,
7203			 "Warning: detected DSPD enabled in EEPROM\n");
7204	}
7205}
7206
7207static netdev_features_t e1000_fix_features(struct net_device *netdev,
7208					    netdev_features_t features)
7209{
7210	struct e1000_adapter *adapter = netdev_priv(netdev);
7211	struct e1000_hw *hw = &adapter->hw;
7212
7213	/* Jumbo frame workaround on 82579 and newer requires CRC be stripped */
7214	if ((hw->mac.type >= e1000_pch2lan) && (netdev->mtu > ETH_DATA_LEN))
7215		features &= ~NETIF_F_RXFCS;
7216
7217	/* Since there is no support for separate Rx/Tx vlan accel
7218	 * enable/disable make sure Tx flag is always in same state as Rx.
7219	 */
7220	if (features & NETIF_F_HW_VLAN_CTAG_RX)
7221		features |= NETIF_F_HW_VLAN_CTAG_TX;
7222	else
7223		features &= ~NETIF_F_HW_VLAN_CTAG_TX;
7224
7225	return features;
7226}
7227
7228static int e1000_set_features(struct net_device *netdev,
7229			      netdev_features_t features)
7230{
7231	struct e1000_adapter *adapter = netdev_priv(netdev);
7232	netdev_features_t changed = features ^ netdev->features;
7233
7234	if (changed & (NETIF_F_TSO | NETIF_F_TSO6))
7235		adapter->flags |= FLAG_TSO_FORCE;
7236
7237	if (!(changed & (NETIF_F_HW_VLAN_CTAG_RX | NETIF_F_HW_VLAN_CTAG_TX |
7238			 NETIF_F_RXCSUM | NETIF_F_RXHASH | NETIF_F_RXFCS |
7239			 NETIF_F_RXALL)))
7240		return 0;
7241
7242	if (changed & NETIF_F_RXFCS) {
7243		if (features & NETIF_F_RXFCS) {
7244			adapter->flags2 &= ~FLAG2_CRC_STRIPPING;
7245		} else {
7246			/* We need to take it back to defaults, which might mean
7247			 * stripping is still disabled at the adapter level.
7248			 */
7249			if (adapter->flags2 & FLAG2_DFLT_CRC_STRIPPING)
7250				adapter->flags2 |= FLAG2_CRC_STRIPPING;
7251			else
7252				adapter->flags2 &= ~FLAG2_CRC_STRIPPING;
7253		}
7254	}
7255
7256	netdev->features = features;
7257
7258	if (netif_running(netdev))
7259		e1000e_reinit_locked(adapter);
7260	else
7261		e1000e_reset(adapter);
7262
7263	return 1;
7264}
7265
7266static const struct net_device_ops e1000e_netdev_ops = {
7267	.ndo_open		= e1000e_open,
7268	.ndo_stop		= e1000e_close,
7269	.ndo_start_xmit		= e1000_xmit_frame,
7270	.ndo_get_stats64	= e1000e_get_stats64,
7271	.ndo_set_rx_mode	= e1000e_set_rx_mode,
7272	.ndo_set_mac_address	= e1000_set_mac,
7273	.ndo_change_mtu		= e1000_change_mtu,
7274	.ndo_do_ioctl		= e1000_ioctl,
7275	.ndo_tx_timeout		= e1000_tx_timeout,
7276	.ndo_validate_addr	= eth_validate_addr,
7277
7278	.ndo_vlan_rx_add_vid	= e1000_vlan_rx_add_vid,
7279	.ndo_vlan_rx_kill_vid	= e1000_vlan_rx_kill_vid,
7280#ifdef CONFIG_NET_POLL_CONTROLLER
7281	.ndo_poll_controller	= e1000_netpoll,
7282#endif
7283	.ndo_set_features = e1000_set_features,
7284	.ndo_fix_features = e1000_fix_features,
7285	.ndo_features_check	= passthru_features_check,
7286};
7287
7288/**
7289 * e1000_probe - Device Initialization Routine
7290 * @pdev: PCI device information struct
7291 * @ent: entry in e1000_pci_tbl
7292 *
7293 * Returns 0 on success, negative on failure
7294 *
7295 * e1000_probe initializes an adapter identified by a pci_dev structure.
7296 * The OS initialization, configuring of the adapter private structure,
7297 * and a hardware reset occur.
7298 **/
7299static int e1000_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
7300{
7301	struct net_device *netdev;
7302	struct e1000_adapter *adapter;
7303	struct e1000_hw *hw;
7304	const struct e1000_info *ei = e1000_info_tbl[ent->driver_data];
7305	resource_size_t mmio_start, mmio_len;
7306	resource_size_t flash_start, flash_len;
7307	static int cards_found;
7308	u16 aspm_disable_flag = 0;
7309	int bars, i, err, pci_using_dac;
7310	u16 eeprom_data = 0;
7311	u16 eeprom_apme_mask = E1000_EEPROM_APME;
7312	s32 ret_val = 0;
7313
7314	if (ei->flags2 & FLAG2_DISABLE_ASPM_L0S)
7315		aspm_disable_flag = PCIE_LINK_STATE_L0S;
7316	if (ei->flags2 & FLAG2_DISABLE_ASPM_L1)
7317		aspm_disable_flag |= PCIE_LINK_STATE_L1;
7318	if (aspm_disable_flag)
7319		e1000e_disable_aspm(pdev, aspm_disable_flag);
7320
7321	err = pci_enable_device_mem(pdev);
7322	if (err)
7323		return err;
7324
7325	pci_using_dac = 0;
7326	err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
7327	if (!err) {
7328		pci_using_dac = 1;
7329	} else {
7330		err = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
7331		if (err) {
7332			dev_err(&pdev->dev,
7333				"No usable DMA configuration, aborting\n");
7334			goto err_dma;
7335		}
7336	}
7337
7338	bars = pci_select_bars(pdev, IORESOURCE_MEM);
7339	err = pci_request_selected_regions_exclusive(pdev, bars,
7340						     e1000e_driver_name);
7341	if (err)
7342		goto err_pci_reg;
7343
7344	/* AER (Advanced Error Reporting) hooks */
7345	pci_enable_pcie_error_reporting(pdev);
7346
7347	pci_set_master(pdev);
7348	/* PCI config space info */
7349	err = pci_save_state(pdev);
7350	if (err)
7351		goto err_alloc_etherdev;
7352
7353	err = -ENOMEM;
7354	netdev = alloc_etherdev(sizeof(struct e1000_adapter));
7355	if (!netdev)
7356		goto err_alloc_etherdev;
7357
7358	SET_NETDEV_DEV(netdev, &pdev->dev);
7359
7360	netdev->irq = pdev->irq;
7361
7362	pci_set_drvdata(pdev, netdev);
7363	adapter = netdev_priv(netdev);
7364	hw = &adapter->hw;
7365	adapter->netdev = netdev;
7366	adapter->pdev = pdev;
7367	adapter->ei = ei;
7368	adapter->pba = ei->pba;
7369	adapter->flags = ei->flags;
7370	adapter->flags2 = ei->flags2;
7371	adapter->hw.adapter = adapter;
7372	adapter->hw.mac.type = ei->mac;
7373	adapter->max_hw_frame_size = ei->max_hw_frame_size;
7374	adapter->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
7375
7376	mmio_start = pci_resource_start(pdev, 0);
7377	mmio_len = pci_resource_len(pdev, 0);
7378
7379	err = -EIO;
7380	adapter->hw.hw_addr = ioremap(mmio_start, mmio_len);
7381	if (!adapter->hw.hw_addr)
7382		goto err_ioremap;
7383
7384	if ((adapter->flags & FLAG_HAS_FLASH) &&
7385	    (pci_resource_flags(pdev, 1) & IORESOURCE_MEM) &&
7386	    (hw->mac.type < e1000_pch_spt)) {
7387		flash_start = pci_resource_start(pdev, 1);
7388		flash_len = pci_resource_len(pdev, 1);
7389		adapter->hw.flash_address = ioremap(flash_start, flash_len);
7390		if (!adapter->hw.flash_address)
7391			goto err_flashmap;
7392	}
7393
7394	/* Set default EEE advertisement */
7395	if (adapter->flags2 & FLAG2_HAS_EEE)
7396		adapter->eee_advert = MDIO_EEE_100TX | MDIO_EEE_1000T;
7397
7398	/* construct the net_device struct */
7399	netdev->netdev_ops = &e1000e_netdev_ops;
7400	e1000e_set_ethtool_ops(netdev);
7401	netdev->watchdog_timeo = 5 * HZ;
7402	netif_napi_add(netdev, &adapter->napi, e1000e_poll, 64);
7403	strlcpy(netdev->name, pci_name(pdev), sizeof(netdev->name));
7404
7405	netdev->mem_start = mmio_start;
7406	netdev->mem_end = mmio_start + mmio_len;
7407
7408	adapter->bd_number = cards_found++;
7409
7410	e1000e_check_options(adapter);
7411
7412	/* setup adapter struct */
7413	err = e1000_sw_init(adapter);
7414	if (err)
7415		goto err_sw_init;
7416
7417	memcpy(&hw->mac.ops, ei->mac_ops, sizeof(hw->mac.ops));
7418	memcpy(&hw->nvm.ops, ei->nvm_ops, sizeof(hw->nvm.ops));
7419	memcpy(&hw->phy.ops, ei->phy_ops, sizeof(hw->phy.ops));
7420
7421	err = ei->get_variants(adapter);
7422	if (err)
7423		goto err_hw_init;
7424
7425	if ((adapter->flags & FLAG_IS_ICH) &&
7426	    (adapter->flags & FLAG_READ_ONLY_NVM) &&
7427	    (hw->mac.type < e1000_pch_spt))
7428		e1000e_write_protect_nvm_ich8lan(&adapter->hw);
7429
7430	hw->mac.ops.get_bus_info(&adapter->hw);
7431
7432	adapter->hw.phy.autoneg_wait_to_complete = 0;
7433
7434	/* Copper options */
7435	if (adapter->hw.phy.media_type == e1000_media_type_copper) {
7436		adapter->hw.phy.mdix = AUTO_ALL_MODES;
7437		adapter->hw.phy.disable_polarity_correction = 0;
7438		adapter->hw.phy.ms_type = e1000_ms_hw_default;
7439	}
7440
7441	if (hw->phy.ops.check_reset_block && hw->phy.ops.check_reset_block(hw))
7442		dev_info(&pdev->dev,
7443			 "PHY reset is blocked due to SOL/IDER session.\n");
7444
7445	/* Set initial default active device features */
7446	netdev->features = (NETIF_F_SG |
7447			    NETIF_F_HW_VLAN_CTAG_RX |
7448			    NETIF_F_HW_VLAN_CTAG_TX |
7449			    NETIF_F_TSO |
7450			    NETIF_F_TSO6 |
7451			    NETIF_F_RXHASH |
7452			    NETIF_F_RXCSUM |
7453			    NETIF_F_HW_CSUM);
7454
7455	/* disable TSO for pcie and 10/100 speeds to avoid
7456	 * some hardware issues and for i219 to fix transfer
7457	 * speed being capped at 60%
7458	 */
7459	if (!(adapter->flags & FLAG_TSO_FORCE)) {
7460		switch (adapter->link_speed) {
7461		case SPEED_10:
7462		case SPEED_100:
7463			e_info("10/100 speed: disabling TSO\n");
7464			netdev->features &= ~NETIF_F_TSO;
7465			netdev->features &= ~NETIF_F_TSO6;
7466			break;
7467		case SPEED_1000:
7468			netdev->features |= NETIF_F_TSO;
7469			netdev->features |= NETIF_F_TSO6;
7470			break;
7471		default:
7472			/* oops */
7473			break;
7474		}
7475		if (hw->mac.type == e1000_pch_spt) {
7476			netdev->features &= ~NETIF_F_TSO;
7477			netdev->features &= ~NETIF_F_TSO6;
7478		}
7479	}
7480
7481	/* Set user-changeable features (subset of all device features) */
7482	netdev->hw_features = netdev->features;
7483	netdev->hw_features |= NETIF_F_RXFCS;
7484	netdev->priv_flags |= IFF_SUPP_NOFCS;
7485	netdev->hw_features |= NETIF_F_RXALL;
7486
7487	if (adapter->flags & FLAG_HAS_HW_VLAN_FILTER)
7488		netdev->features |= NETIF_F_HW_VLAN_CTAG_FILTER;
7489
7490	netdev->vlan_features |= (NETIF_F_SG |
7491				  NETIF_F_TSO |
7492				  NETIF_F_TSO6 |
7493				  NETIF_F_HW_CSUM);
7494
7495	netdev->priv_flags |= IFF_UNICAST_FLT;
7496
7497	if (pci_using_dac) {
7498		netdev->features |= NETIF_F_HIGHDMA;
7499		netdev->vlan_features |= NETIF_F_HIGHDMA;
7500	}
7501
7502	/* MTU range: 68 - max_hw_frame_size */
7503	netdev->min_mtu = ETH_MIN_MTU;
7504	netdev->max_mtu = adapter->max_hw_frame_size -
7505			  (VLAN_ETH_HLEN + ETH_FCS_LEN);
7506
7507	if (e1000e_enable_mng_pass_thru(&adapter->hw))
7508		adapter->flags |= FLAG_MNG_PT_ENABLED;
7509
7510	/* before reading the NVM, reset the controller to
7511	 * put the device in a known good starting state
7512	 */
7513	adapter->hw.mac.ops.reset_hw(&adapter->hw);
7514
7515	/* systems with ASPM and others may see the checksum fail on the first
7516	 * attempt. Let's give it a few tries
7517	 */
7518	for (i = 0;; i++) {
7519		if (e1000_validate_nvm_checksum(&adapter->hw) >= 0)
7520			break;
7521		if (i == 2) {
7522			dev_err(&pdev->dev, "The NVM Checksum Is Not Valid\n");
7523			err = -EIO;
7524			goto err_eeprom;
7525		}
7526	}
7527
7528	e1000_eeprom_checks(adapter);
7529
7530	/* copy the MAC address */
7531	if (e1000e_read_mac_addr(&adapter->hw))
7532		dev_err(&pdev->dev,
7533			"NVM Read Error while reading MAC address\n");
7534
7535	memcpy(netdev->dev_addr, adapter->hw.mac.addr, netdev->addr_len);
7536
7537	if (!is_valid_ether_addr(netdev->dev_addr)) {
7538		dev_err(&pdev->dev, "Invalid MAC Address: %pM\n",
7539			netdev->dev_addr);
7540		err = -EIO;
7541		goto err_eeprom;
7542	}
7543
7544	timer_setup(&adapter->watchdog_timer, e1000_watchdog, 0);
7545	timer_setup(&adapter->phy_info_timer, e1000_update_phy_info, 0);
7546
7547	INIT_WORK(&adapter->reset_task, e1000_reset_task);
7548	INIT_WORK(&adapter->watchdog_task, e1000_watchdog_task);
7549	INIT_WORK(&adapter->downshift_task, e1000e_downshift_workaround);
7550	INIT_WORK(&adapter->update_phy_task, e1000e_update_phy_task);
7551	INIT_WORK(&adapter->print_hang_task, e1000_print_hw_hang);
7552
7553	/* Initialize link parameters. User can change them with ethtool */
7554	adapter->hw.mac.autoneg = 1;
7555	adapter->fc_autoneg = true;
7556	adapter->hw.fc.requested_mode = e1000_fc_default;
7557	adapter->hw.fc.current_mode = e1000_fc_default;
7558	adapter->hw.phy.autoneg_advertised = 0x2f;
7559
7560	/* Initial Wake on LAN setting - If APM wake is enabled in
7561	 * the EEPROM, enable the ACPI Magic Packet filter
7562	 */
7563	if (adapter->flags & FLAG_APME_IN_WUC) {
7564		/* APME bit in EEPROM is mapped to WUC.APME */
7565		eeprom_data = er32(WUC);
7566		eeprom_apme_mask = E1000_WUC_APME;
7567		if ((hw->mac.type > e1000_ich10lan) &&
7568		    (eeprom_data & E1000_WUC_PHY_WAKE))
7569			adapter->flags2 |= FLAG2_HAS_PHY_WAKEUP;
7570	} else if (adapter->flags & FLAG_APME_IN_CTRL3) {
7571		if (adapter->flags & FLAG_APME_CHECK_PORT_B &&
7572		    (adapter->hw.bus.func == 1))
7573			ret_val = e1000_read_nvm(&adapter->hw,
7574					      NVM_INIT_CONTROL3_PORT_B,
7575					      1, &eeprom_data);
7576		else
7577			ret_val = e1000_read_nvm(&adapter->hw,
7578					      NVM_INIT_CONTROL3_PORT_A,
7579					      1, &eeprom_data);
7580	}
7581
7582	/* fetch WoL from EEPROM */
7583	if (ret_val)
7584		e_dbg("NVM read error getting WoL initial values: %d\n", ret_val);
7585	else if (eeprom_data & eeprom_apme_mask)
7586		adapter->eeprom_wol |= E1000_WUFC_MAG;
7587
7588	/* now that we have the eeprom settings, apply the special cases
7589	 * where the eeprom may be wrong or the board simply won't support
7590	 * wake on lan on a particular port
7591	 */
7592	if (!(adapter->flags & FLAG_HAS_WOL))
7593		adapter->eeprom_wol = 0;
7594
7595	/* initialize the wol settings based on the eeprom settings */
7596	adapter->wol = adapter->eeprom_wol;
7597
7598	/* make sure adapter isn't asleep if manageability is enabled */
7599	if (adapter->wol || (adapter->flags & FLAG_MNG_PT_ENABLED) ||
7600	    (hw->mac.ops.check_mng_mode(hw)))
7601		device_wakeup_enable(&pdev->dev);
7602
7603	/* save off EEPROM version number */
7604	ret_val = e1000_read_nvm(&adapter->hw, 5, 1, &adapter->eeprom_vers);
7605
7606	if (ret_val) {
7607		e_dbg("NVM read error getting EEPROM version: %d\n", ret_val);
7608		adapter->eeprom_vers = 0;
7609	}
7610
7611	/* init PTP hardware clock */
7612	e1000e_ptp_init(adapter);
7613
7614	/* reset the hardware with the new settings */
7615	e1000e_reset(adapter);
7616
7617	/* If the controller has AMT, do not set DRV_LOAD until the interface
7618	 * is up.  For all other cases, let the f/w know that the h/w is now
7619	 * under the control of the driver.
7620	 */
7621	if (!(adapter->flags & FLAG_HAS_AMT))
7622		e1000e_get_hw_control(adapter);
7623
7624	if (hw->mac.type >= e1000_pch_cnp)
7625		adapter->flags2 |= FLAG2_ENABLE_S0IX_FLOWS;
7626
7627	strlcpy(netdev->name, "eth%d", sizeof(netdev->name));
7628	err = register_netdev(netdev);
7629	if (err)
7630		goto err_register;
7631
7632	/* carrier off reporting is important to ethtool even BEFORE open */
7633	netif_carrier_off(netdev);
7634
7635	e1000_print_device_info(adapter);
7636
7637	dev_pm_set_driver_flags(&pdev->dev, DPM_FLAG_NO_DIRECT_COMPLETE);
7638
7639	if (pci_dev_run_wake(pdev) && hw->mac.type < e1000_pch_cnp)
7640		pm_runtime_put_noidle(&pdev->dev);
7641
7642	return 0;
7643
7644err_register:
7645	if (!(adapter->flags & FLAG_HAS_AMT))
7646		e1000e_release_hw_control(adapter);
7647err_eeprom:
7648	if (hw->phy.ops.check_reset_block && !hw->phy.ops.check_reset_block(hw))
7649		e1000_phy_hw_reset(&adapter->hw);
7650err_hw_init:
7651	kfree(adapter->tx_ring);
7652	kfree(adapter->rx_ring);
7653err_sw_init:
7654	if ((adapter->hw.flash_address) && (hw->mac.type < e1000_pch_spt))
7655		iounmap(adapter->hw.flash_address);
7656	e1000e_reset_interrupt_capability(adapter);
7657err_flashmap:
7658	iounmap(adapter->hw.hw_addr);
7659err_ioremap:
7660	free_netdev(netdev);
7661err_alloc_etherdev:
7662	pci_disable_pcie_error_reporting(pdev);
7663	pci_release_mem_regions(pdev);
7664err_pci_reg:
7665err_dma:
7666	pci_disable_device(pdev);
7667	return err;
7668}
7669
7670/**
7671 * e1000_remove - Device Removal Routine
7672 * @pdev: PCI device information struct
7673 *
7674 * e1000_remove is called by the PCI subsystem to alert the driver
7675 * that it should release a PCI device.  The could be caused by a
7676 * Hot-Plug event, or because the driver is going to be removed from
7677 * memory.
7678 **/
7679static void e1000_remove(struct pci_dev *pdev)
7680{
7681	struct net_device *netdev = pci_get_drvdata(pdev);
7682	struct e1000_adapter *adapter = netdev_priv(netdev);
7683
7684	e1000e_ptp_remove(adapter);
7685
7686	/* The timers may be rescheduled, so explicitly disable them
7687	 * from being rescheduled.
7688	 */
7689	set_bit(__E1000_DOWN, &adapter->state);
7690	del_timer_sync(&adapter->watchdog_timer);
7691	del_timer_sync(&adapter->phy_info_timer);
7692
7693	cancel_work_sync(&adapter->reset_task);
7694	cancel_work_sync(&adapter->watchdog_task);
7695	cancel_work_sync(&adapter->downshift_task);
7696	cancel_work_sync(&adapter->update_phy_task);
7697	cancel_work_sync(&adapter->print_hang_task);
7698
7699	if (adapter->flags & FLAG_HAS_HW_TIMESTAMP) {
7700		cancel_work_sync(&adapter->tx_hwtstamp_work);
7701		if (adapter->tx_hwtstamp_skb) {
7702			dev_consume_skb_any(adapter->tx_hwtstamp_skb);
7703			adapter->tx_hwtstamp_skb = NULL;
7704		}
7705	}
7706
7707	unregister_netdev(netdev);
7708
7709	if (pci_dev_run_wake(pdev))
7710		pm_runtime_get_noresume(&pdev->dev);
7711
7712	/* Release control of h/w to f/w.  If f/w is AMT enabled, this
7713	 * would have already happened in close and is redundant.
7714	 */
7715	e1000e_release_hw_control(adapter);
7716
7717	e1000e_reset_interrupt_capability(adapter);
7718	kfree(adapter->tx_ring);
7719	kfree(adapter->rx_ring);
7720
7721	iounmap(adapter->hw.hw_addr);
7722	if ((adapter->hw.flash_address) &&
7723	    (adapter->hw.mac.type < e1000_pch_spt))
7724		iounmap(adapter->hw.flash_address);
7725	pci_release_mem_regions(pdev);
7726
7727	free_netdev(netdev);
7728
7729	/* AER disable */
7730	pci_disable_pcie_error_reporting(pdev);
7731
7732	pci_disable_device(pdev);
7733}
7734
7735/* PCI Error Recovery (ERS) */
7736static const struct pci_error_handlers e1000_err_handler = {
7737	.error_detected = e1000_io_error_detected,
7738	.slot_reset = e1000_io_slot_reset,
7739	.resume = e1000_io_resume,
7740};
7741
7742static const struct pci_device_id e1000_pci_tbl[] = {
7743	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_COPPER), board_82571 },
7744	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_FIBER), board_82571 },
7745	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_QUAD_COPPER), board_82571 },
7746	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_QUAD_COPPER_LP),
7747	  board_82571 },
7748	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_QUAD_FIBER), board_82571 },
7749	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_SERDES), board_82571 },
7750	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_SERDES_DUAL), board_82571 },
7751	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571EB_SERDES_QUAD), board_82571 },
7752	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82571PT_QUAD_COPPER), board_82571 },
7753
7754	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82572EI), board_82572 },
7755	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82572EI_COPPER), board_82572 },
7756	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82572EI_FIBER), board_82572 },
7757	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82572EI_SERDES), board_82572 },
7758
7759	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82573E), board_82573 },
7760	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82573E_IAMT), board_82573 },
7761	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82573L), board_82573 },
7762
7763	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82574L), board_82574 },
7764	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82574LA), board_82574 },
7765	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_82583V), board_82583 },
7766
7767	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_80003ES2LAN_COPPER_DPT),
7768	  board_80003es2lan },
7769	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_80003ES2LAN_COPPER_SPT),
7770	  board_80003es2lan },
7771	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_80003ES2LAN_SERDES_DPT),
7772	  board_80003es2lan },
7773	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_80003ES2LAN_SERDES_SPT),
7774	  board_80003es2lan },
7775
7776	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH8_IFE), board_ich8lan },
7777	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH8_IFE_G), board_ich8lan },
7778	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH8_IFE_GT), board_ich8lan },
7779	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH8_IGP_AMT), board_ich8lan },
7780	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH8_IGP_C), board_ich8lan },
7781	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH8_IGP_M), board_ich8lan },
7782	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH8_IGP_M_AMT), board_ich8lan },
7783	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH8_82567V_3), board_ich8lan },
7784
7785	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IFE), board_ich9lan },
7786	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IFE_G), board_ich9lan },
7787	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IFE_GT), board_ich9lan },
7788	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_AMT), board_ich9lan },
7789	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_C), board_ich9lan },
7790	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_BM), board_ich9lan },
7791	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M), board_ich9lan },
7792	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_AMT), board_ich9lan },
7793	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_V), board_ich9lan },
7794
7795	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LM), board_ich9lan },
7796	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LF), board_ich9lan },
7797	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_V), board_ich9lan },
7798
7799	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_D_BM_LM), board_ich10lan },
7800	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_D_BM_LF), board_ich10lan },
7801	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_D_BM_V), board_ich10lan },
7802
7803	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_M_HV_LM), board_pchlan },
7804	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_M_HV_LC), board_pchlan },
7805	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_D_HV_DM), board_pchlan },
7806	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_D_HV_DC), board_pchlan },
7807
7808	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH2_LV_LM), board_pch2lan },
7809	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH2_LV_V), board_pch2lan },
7810
7811	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LPT_I217_LM), board_pch_lpt },
7812	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LPT_I217_V), board_pch_lpt },
7813	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LPTLP_I218_LM), board_pch_lpt },
7814	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LPTLP_I218_V), board_pch_lpt },
7815	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_I218_LM2), board_pch_lpt },
7816	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_I218_V2), board_pch_lpt },
7817	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_I218_LM3), board_pch_lpt },
7818	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_I218_V3), board_pch_lpt },
7819	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_LM), board_pch_spt },
7820	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_V), board_pch_spt },
7821	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_LM2), board_pch_spt },
7822	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_V2), board_pch_spt },
7823	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_LBG_I219_LM3), board_pch_spt },
7824	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_LM4), board_pch_spt },
7825	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_V4), board_pch_spt },
7826	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_LM5), board_pch_spt },
7827	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_SPT_I219_V5), board_pch_spt },
7828	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CNP_I219_LM6), board_pch_cnp },
7829	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CNP_I219_V6), board_pch_cnp },
7830	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CNP_I219_LM7), board_pch_cnp },
7831	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CNP_I219_V7), board_pch_cnp },
7832	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ICP_I219_LM8), board_pch_cnp },
7833	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ICP_I219_V8), board_pch_cnp },
7834	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ICP_I219_LM9), board_pch_cnp },
7835	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ICP_I219_V9), board_pch_cnp },
7836	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_LM10), board_pch_cnp },
7837	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_V10), board_pch_cnp },
7838	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_LM11), board_pch_cnp },
7839	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_V11), board_pch_cnp },
7840	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_LM12), board_pch_spt },
7841	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_CMP_I219_V12), board_pch_spt },
7842	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_LM13), board_pch_tgp },
7843	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_V13), board_pch_tgp },
7844	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_LM14), board_pch_tgp },
7845	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_V14), board_pch_tgp },
7846	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_LM15), board_pch_tgp },
7847	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_TGP_I219_V15), board_pch_tgp },
7848	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM16), board_pch_tgp },
7849	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V16), board_pch_tgp },
7850	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_LM17), board_pch_tgp },
7851	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_ADP_I219_V17), board_pch_tgp },
7852	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_LM18), board_pch_tgp },
7853	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_V18), board_pch_tgp },
7854	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_LM19), board_pch_tgp },
7855	{ PCI_VDEVICE(INTEL, E1000_DEV_ID_PCH_MTP_I219_V19), board_pch_tgp },
7856
7857	{ 0, 0, 0, 0, 0, 0, 0 }	/* terminate list */
7858};
7859MODULE_DEVICE_TABLE(pci, e1000_pci_tbl);
7860
7861static const struct dev_pm_ops e1000_pm_ops = {
7862#ifdef CONFIG_PM_SLEEP
7863	.suspend	= e1000e_pm_suspend,
7864	.resume		= e1000e_pm_resume,
7865	.freeze		= e1000e_pm_freeze,
7866	.thaw		= e1000e_pm_thaw,
7867	.poweroff	= e1000e_pm_suspend,
7868	.restore	= e1000e_pm_resume,
7869#endif
7870	SET_RUNTIME_PM_OPS(e1000e_pm_runtime_suspend, e1000e_pm_runtime_resume,
7871			   e1000e_pm_runtime_idle)
7872};
7873
7874/* PCI Device API Driver */
7875static struct pci_driver e1000_driver = {
7876	.name     = e1000e_driver_name,
7877	.id_table = e1000_pci_tbl,
7878	.probe    = e1000_probe,
7879	.remove   = e1000_remove,
7880	.driver   = {
7881		.pm = &e1000_pm_ops,
7882	},
7883	.shutdown = e1000_shutdown,
7884	.err_handler = &e1000_err_handler
7885};
7886
7887/**
7888 * e1000_init_module - Driver Registration Routine
7889 *
7890 * e1000_init_module is the first routine called when the driver is
7891 * loaded. All it does is register with the PCI subsystem.
7892 **/
7893static int __init e1000_init_module(void)
7894{
7895	pr_info("Intel(R) PRO/1000 Network Driver\n");
7896	pr_info("Copyright(c) 1999 - 2015 Intel Corporation.\n");
7897
7898	return pci_register_driver(&e1000_driver);
7899}
7900module_init(e1000_init_module);
7901
7902/**
7903 * e1000_exit_module - Driver Exit Cleanup Routine
7904 *
7905 * e1000_exit_module is called just before the driver is removed
7906 * from memory.
7907 **/
7908static void __exit e1000_exit_module(void)
7909{
7910	pci_unregister_driver(&e1000_driver);
7911}
7912module_exit(e1000_exit_module);
7913
7914MODULE_AUTHOR("Intel Corporation, <linux.nics@intel.com>");
7915MODULE_DESCRIPTION("Intel(R) PRO/1000 Network Driver");
7916MODULE_LICENSE("GPL v2");
7917
7918/* netdev.c */
7919