Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

Pull networking updates from Jakub Kicinski:
 "Core:

   - Add dedicated kmem_cache for typical/small skb->head, avoid having
     to access struct page at kfree time, and improve memory use.

   - Introduce sysctl to set default RPS configuration for new netdevs.

   - Define Netlink protocol specification format which can be used to
     describe messages used by each family and auto-generate parsers.
     Add tools for generating kernel data structures and uAPI headers.

   - Expose all net/core sysctls inside netns.

   - Remove 4s sleep in netpoll if carrier is instantly detected on
     boot.

   - Add configurable limit of MDB entries per port, and port-vlan.

   - Continue populating drop reasons throughout the stack.

   - Retire a handful of legacy Qdiscs and classifiers.

  Protocols:

   - Support IPv4 big TCP (TSO frames larger than 64kB).

   - Add IP_LOCAL_PORT_RANGE socket option, to control local port range
     on socket by socket basis.

   - Track and report in procfs number of MPTCP sockets used.

   - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path
     manager.

   - IPv6: don't check net.ipv6.route.max_size and rely on garbage
     collection to free memory (similarly to IPv4).

   - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986).

   - ICMP: add per-rate limit counters.

   - Add support for user scanning requests in ieee802154.

   - Remove static WEP support.

   - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate
     reporting.

   - WiFi 7 EHT channel puncturing support (client & AP).

  BPF:

   - Add a rbtree data structure following the "next-gen data structure"
     precedent set by recently added linked list, that is, by using
     kfunc + kptr instead of adding a new BPF map type.

   - Expose XDP hints via kfuncs with initial support for RX hash and
     timestamp metadata.

   - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to
     better support decap on GRE tunnel devices not operating in collect
     metadata.

   - Improve x86 JIT's codegen for PROBE_MEM runtime error checks.

   - Remove the need for trace_printk_lock for bpf_trace_printk and
     bpf_trace_vprintk helpers.

   - Extend libbpf's bpf_tracing.h support for tracing arguments of
     kprobes/uprobes and syscall as a special case.

   - Significantly reduce the search time for module symbols by
     livepatch and BPF.

   - Enable cpumasks to be used as kptrs, which is useful for tracing
     programs tracking which tasks end up running on which CPUs in
     different time intervals.

   - Add support for BPF trampoline on s390x and riscv64.

   - Add capability to export the XDP features supported by the NIC.

   - Add __bpf_kfunc tag for marking kernel functions as kfuncs.

   - Add cgroup.memory=nobpf kernel parameter option to disable BPF
     memory accounting for container environments.

  Netfilter:

   - Remove the CLUSTERIP target. It has been marked as obsolete for
     years, and we still have WARN splats wrt races of the out-of-band
     /proc interface installed by this target.

   - Add 'destroy' commands to nf_tables. They are identical to the
     existing 'delete' commands, but do not return an error if the
     referenced object (set, chain, rule...) did not exist.

  Driver API:

   - Improve cpumask_local_spread() locality to help NICs set the right
     IRQ affinity on AMD platforms.

   - Separate C22 and C45 MDIO bus transactions more clearly.

   - Introduce new DCB table to control DSCP rewrite on egress.

   - Support configuration of Physical Layer Collision Avoidance (PLCA)
     Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of
     shared medium Ethernet.

   - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing
     preemption of low priority frames by high priority frames.

   - Add support for controlling MACSec offload using netlink SET.

   - Rework devlink instance refcounts to allow registration and
     de-registration under the instance lock. Split the code into
     multiple files, drop some of the unnecessarily granular locks and
     factor out common parts of netlink operation handling.

   - Add TX frame aggregation parameters (for USB drivers).

   - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning
     messages with notifications for debug.

   - Allow offloading of UDP NEW connections via act_ct.

   - Add support for per action HW stats in TC.

   - Support hardware miss to TC action (continue processing in SW from
     a specific point in the action chain).

   - Warn if old Wireless Extension user space interface is used with
     modern cfg80211/mac80211 drivers. Do not support Wireless
     Extensions for Wi-Fi 7 devices at all. Everyone should switch to
     using nl80211 interface instead.

   - Improve the CAN bit timing configuration. Use extack to return
     error messages directly to user space, update the SJW handling,
     including the definition of a new default value that will benefit
     CAN-FD controllers, by increasing their oscillator tolerance.

  New hardware / drivers:

   - Ethernet:
      - nVidia BlueField-3 support (control traffic driver)
      - Ethernet support for imx93 SoCs
      - Motorcomm yt8531 gigabit Ethernet PHY
      - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA)
      - Microchip LAN8841 PHY (incl. cable diagnostics and PTP)
      - Amlogic gxl MDIO mux

   - WiFi:
      - RealTek RTL8188EU (rtl8xxxu)
      - Qualcomm Wi-Fi 7 devices (ath12k)

   - CAN:
      - Renesas R-Car V4H

  Drivers:

   - Bluetooth:
      - Set Per Platform Antenna Gain (PPAG) for Intel controllers.

   - Ethernet NICs:
      - Intel (1G, igc):
         - support TSN / Qbv / packet scheduling features of i226 model
      - Intel (100G, ice):
         - use GNSS subsystem instead of TTY
         - multi-buffer XDP support
         - extend support for GPIO pins to E823 devices
      - nVidia/Mellanox:
         - update the shared buffer configuration on PFC commands
         - implement PTP adjphase function for HW offset control
         - TC support for Geneve and GRE with VF tunnel offload
         - more efficient crypto key management method
         - multi-port eswitch support
      - Netronome/Corigine:
         - add DCB IEEE support
         - support IPsec offloading for NFP3800
      - Freescale/NXP (enetc):
         - support XDP_REDIRECT for XDP non-linear buffers
         - improve reconfig, avoid link flap and waiting for idle
         - support MAC Merge layer
      - Other NICs:
         - sfc/ef100: add basic devlink support for ef100
         - ionic: rx_push mode operation (writing descriptors via MMIO)
         - bnxt: use the auxiliary bus abstraction for RDMA
         - r8169: disable ASPM and reset bus in case of tx timeout
         - cpsw: support QSGMII mode for J721e CPSW9G
         - cpts: support pulse-per-second output
         - ngbe: add an mdio bus driver
         - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing
         - r8152: handle devices with FW with NCM support
         - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation
         - virtio-net: support multi buffer XDP
         - virtio/vsock: replace virtio_vsock_pkt with sk_buff
         - tsnep: XDP support

   - Ethernet high-speed switches:
      - nVidia/Mellanox (mlxsw):
         - add support for latency TLV (in FW control messages)
      - Microchip (sparx5):
         - separate explicit and implicit traffic forwarding rules, make
           the implicit rules always active
         - add support for egress DSCP rewrite
         - IS0 VCAP support (Ingress Classification)
         - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS
           etc.)
         - ES2 VCAP support (Egress Access Control)
         - support for Per-Stream Filtering and Policing (802.1Q,
           8.6.5.1)

   - Ethernet embedded switches:
      - Marvell (mv88e6xxx):
         - add MAB (port auth) offload support
         - enable PTP receive for mv88e6390
      - NXP (ocelot):
         - support MAC Merge layer
         - support for the the vsc7512 internal copper phys
      - Microchip:
         - lan9303: convert to PHYLINK
         - lan966x: support TC flower filter statistics
         - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x
         - lan937x: support Credit Based Shaper configuration
         - ksz9477: support Energy Efficient Ethernet
      - other:
         - qca8k: convert to regmap read/write API, use bulk operations
         - rswitch: Improve TX timestamp accuracy

   - Intel WiFi (iwlwifi):
      - EHT (Wi-Fi 7) rate reporting
      - STEP equalizer support: transfer some STEP (connection to radio
        on platforms with integrated wifi) related parameters from the
        BIOS to the firmware.

   - Qualcomm 802.11ax WiFi (ath11k):
      - IPQ5018 support
      - Fine Timing Measurement (FTM) responder role support
      - channel 177 support

   - MediaTek WiFi (mt76):
      - per-PHY LED support
      - mt7996: EHT (Wi-Fi 7) support
      - Wireless Ethernet Dispatch (WED) reset support
      - switch to using page pool allocator

   - RealTek WiFi (rtw89):
      - support new version of Bluetooth co-existance

   - Mobile:
      - rmnet: support TX aggregation"

* tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits)
  page_pool: add a comment explaining the fragment counter usage
  net: ethtool: fix __ethtool_dev_mm_supported() implementation
  ethtool: pse-pd: Fix double word in comments
  xsk: add linux/vmalloc.h to xsk.c
  sefltests: netdevsim: wait for devlink instance after netns removal
  selftest: fib_tests: Always cleanup before exit
  net/mlx5e: Align IPsec ASO result memory to be as required by hardware
  net/mlx5e: TC, Set CT miss to the specific ct action instance
  net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG
  net/mlx5: Refactor tc miss handling to a single function
  net/mlx5: Kconfig: Make tc offload depend on tc skb extension
  net/sched: flower: Support hardware miss to tc action
  net/sched: flower: Move filter handle initialization earlier
  net/sched: cls_api: Support hardware miss to tc action
  net/sched: Rename user cookie and act cookie
  sfc: fix builds without CONFIG_RTC_LIB
  sfc: clean up some inconsistent indentings
  net/mlx4_en: Introduce flexible array to silence overflow warning
  net: lan966x: Fix possible deadlock inside PTP
  net/ulp: Remove redundant ->clone() test in inet_clone_ulp().
  ...

1 files changed, 734 insertions, 0 deletions

diff --git a/drivers/block/paride/pg.c b/drivers/block/paride/pg.c
new file mode 100644
index 000000000..3b5882bfb
--- /dev/null
+++ b/drivers/block/paride/pg.c
@@ -0,0 +1,734 @@
+/* 
+	pg.c    (c) 1998  Grant R. Guenther <grant@torque.net>
+			  Under the terms of the GNU General Public License.
+
+	The pg driver provides a simple character device interface for
+	sending ATAPI commands to a device.  With the exception of the
+	ATAPI reset operation, all operations are performed by a pair
+	of read and write operations to the appropriate /dev/pgN device.
+	A write operation delivers a command and any outbound data in
+	a single buffer.  Normally, the write will succeed unless the
+	device is offline or malfunctioning, or there is already another
+	command pending.  If the write succeeds, it should be followed
+	immediately by a read operation, to obtain any returned data and
+	status information.  A read will fail if there is no operation
+	in progress.
+
+	As a special case, the device can be reset with a write operation,
+	and in this case, no following read is expected, or permitted.
+
+	There are no ioctl() operations.  Any single operation
+	may transfer at most PG_MAX_DATA bytes.  Note that the driver must
+	copy the data through an internal buffer.  In keeping with all
+	current ATAPI devices, command packets are assumed to be exactly
+	12 bytes in length.
+
+	To permit future changes to this interface, the headers in the
+	read and write buffers contain a single character "magic" flag.
+	Currently this flag must be the character "P".
+
+	By default, the driver will autoprobe for a single parallel
+	port ATAPI device, but if their individual parameters are
+	specified, the driver can handle up to 4 devices.
+
+	To use this device, you must have the following device 
+	special files defined:
+
+		/dev/pg0 c 97 0
+		/dev/pg1 c 97 1
+		/dev/pg2 c 97 2
+		/dev/pg3 c 97 3
+
+	(You'll need to change the 97 to something else if you use
+	the 'major' parameter to install the driver on a different
+	major number.)
+
+	The behaviour of the pg driver can be altered by setting
+	some parameters from the insmod command line.  The following
+	parameters are adjustable:
+
+	    drive0      These four arguments can be arrays of       
+	    drive1      1-6 integers as follows:
+	    drive2
+	    drive3      <prt>,<pro>,<uni>,<mod>,<slv>,<dly>
+
+			Where,
+
+		<prt>   is the base of the parallel port address for
+			the corresponding drive.  (required)
+
+		<pro>   is the protocol number for the adapter that
+			supports this drive.  These numbers are
+			logged by 'paride' when the protocol modules
+			are initialised.  (0 if not given)
+
+		<uni>   for those adapters that support chained
+			devices, this is the unit selector for the
+			chain of devices on the given port.  It should
+			be zero for devices that don't support chaining.
+			(0 if not given)
+
+		<mod>   this can be -1 to choose the best mode, or one
+			of the mode numbers supported by the adapter.
+			(-1 if not given)
+
+		<slv>   ATAPI devices can be jumpered to master or slave.
+			Set this to 0 to choose the master drive, 1 to
+			choose the slave, -1 (the default) to choose the
+			first drive found.
+
+		<dly>   some parallel ports require the driver to 
+			go more slowly.  -1 sets a default value that
+			should work with the chosen protocol.  Otherwise,
+			set this to a small integer, the larger it is
+			the slower the port i/o.  In some cases, setting
+			this to zero will speed up the device. (default -1)
+
+	    major	You may use this parameter to override the
+			default major number (97) that this driver
+			will use.  Be sure to change the device
+			name as well.
+
+	    name	This parameter is a character string that
+			contains the name the kernel will use for this
+			device (in /proc output, for instance).
+			(default "pg").
+
+	    verbose     This parameter controls the amount of logging
+			that is done by the driver.  Set it to 0 for 
+			quiet operation, to 1 to enable progress
+			messages while the driver probes for devices,
+			or to 2 for full debug logging.  (default 0)
+
+	If this driver is built into the kernel, you can use 
+	the following command line parameters, with the same values
+	as the corresponding module parameters listed above:
+
+	    pg.drive0
+	    pg.drive1
+	    pg.drive2
+	    pg.drive3
+
+	In addition, you can use the parameter pg.disable to disable
+	the driver entirely.
+
+*/
+
+/* Changes:
+
+	1.01	GRG 1998.06.16	Bug fixes
+	1.02    GRG 1998.09.24  Added jumbo support
+
+*/
+
+#define PG_VERSION      "1.02"
+#define PG_MAJOR	97
+#define PG_NAME		"pg"
+#define PG_UNITS	4
+
+#ifndef PI_PG
+#define PI_PG	4
+#endif
+
+#include <linux/types.h>
+/* Here are things one can override from the insmod command.
+   Most are autoprobed by paride unless set here.  Verbose is 0
+   by default.
+
+*/
+
+static int verbose;
+static int major = PG_MAJOR;
+static char *name = PG_NAME;
+static int disable = 0;
+
+static int drive0[6] = { 0, 0, 0, -1, -1, -1 };
+static int drive1[6] = { 0, 0, 0, -1, -1, -1 };
+static int drive2[6] = { 0, 0, 0, -1, -1, -1 };
+static int drive3[6] = { 0, 0, 0, -1, -1, -1 };
+
+static int (*drives[4])[6] = {&drive0, &drive1, &drive2, &drive3};
+static int pg_drive_count;
+
+enum {D_PRT, D_PRO, D_UNI, D_MOD, D_SLV, D_DLY};
+
+/* end of parameters */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/fs.h>
+#include <linux/delay.h>
+#include <linux/slab.h>
+#include <linux/mtio.h>
+#include <linux/pg.h>
+#include <linux/device.h>
+#include <linux/sched.h>	/* current, TASK_* */
+#include <linux/mutex.h>
+#include <linux/jiffies.h>
+
+#include <linux/uaccess.h>
+
+module_param(verbose, int, 0644);
+module_param(major, int, 0);
+module_param(name, charp, 0);
+module_param_array(drive0, int, NULL, 0);
+module_param_array(drive1, int, NULL, 0);
+module_param_array(drive2, int, NULL, 0);
+module_param_array(drive3, int, NULL, 0);
+
+#include "paride.h"
+
+#define PG_SPIN_DEL     50	/* spin delay in micro-seconds  */
+#define PG_SPIN         200
+#define PG_TMO		HZ
+#define PG_RESET_TMO	10*HZ
+
+#define STAT_ERR        0x01
+#define STAT_INDEX      0x02
+#define STAT_ECC        0x04
+#define STAT_DRQ        0x08
+#define STAT_SEEK       0x10
+#define STAT_WRERR      0x20
+#define STAT_READY      0x40
+#define STAT_BUSY       0x80
+
+#define ATAPI_IDENTIFY		0x12
+
+static DEFINE_MUTEX(pg_mutex);
+static int pg_open(struct inode *inode, struct file *file);
+static int pg_release(struct inode *inode, struct file *file);
+static ssize_t pg_read(struct file *filp, char __user *buf,
+		       size_t count, loff_t * ppos);
+static ssize_t pg_write(struct file *filp, const char __user *buf,
+			size_t count, loff_t * ppos);
+static int pg_detect(void);
+
+#define PG_NAMELEN      8
+
+struct pg {
+	struct pi_adapter pia;	/* interface to paride layer */
+	struct pi_adapter *pi;
+	int busy;		/* write done, read expected */
+	int start;		/* jiffies at command start */
+	int dlen;		/* transfer size requested */
+	unsigned long timeout;	/* timeout requested */
+	int status;		/* last sense key */
+	int drive;		/* drive */
+	unsigned long access;	/* count of active opens ... */
+	int present;		/* device present ? */
+	char *bufptr;
+	char name[PG_NAMELEN];	/* pg0, pg1, ... */
+};
+
+static struct pg devices[PG_UNITS];
+
+static int pg_identify(struct pg *dev, int log);
+
+static char pg_scratch[512];	/* scratch block buffer */
+
+static struct class *pg_class;
+static void *par_drv;		/* reference of parport driver */
+
+/* kernel glue structures */
+
+static const struct file_operations pg_fops = {
+	.owner = THIS_MODULE,
+	.read = pg_read,
+	.write = pg_write,
+	.open = pg_open,
+	.release = pg_release,
+	.llseek = noop_llseek,
+};
+
+static void pg_init_units(void)
+{
+	int unit;
+
+	pg_drive_count = 0;
+	for (unit = 0; unit < PG_UNITS; unit++) {
+		int *parm = *drives[unit];
+		struct pg *dev = &devices[unit];
+		dev->pi = &dev->pia;
+		clear_bit(0, &dev->access);
+		dev->busy = 0;
+		dev->present = 0;
+		dev->bufptr = NULL;
+		dev->drive = parm[D_SLV];
+		snprintf(dev->name, PG_NAMELEN, "%s%c", name, 'a'+unit);
+		if (parm[D_PRT])
+			pg_drive_count++;
+	}
+}
+
+static inline int status_reg(struct pg *dev)
+{
+	return pi_read_regr(dev->pi, 1, 6);
+}
+
+static inline int read_reg(struct pg *dev, int reg)
+{
+	return pi_read_regr(dev->pi, 0, reg);
+}
+
+static inline void write_reg(struct pg *dev, int reg, int val)
+{
+	pi_write_regr(dev->pi, 0, reg, val);
+}
+
+static inline u8 DRIVE(struct pg *dev)
+{
+	return 0xa0+0x10*dev->drive;
+}
+
+static void pg_sleep(int cs)
+{
+	schedule_timeout_interruptible(cs);
+}
+
+static int pg_wait(struct pg *dev, int go, int stop, unsigned long tmo, char *msg)
+{
+	int j, r, e, s, p, to;
+
+	dev->status = 0;
+
+	j = 0;
+	while ((((r = status_reg(dev)) & go) || (stop && (!(r & stop))))
+	       && time_before(jiffies, tmo)) {
+		if (j++ < PG_SPIN)
+			udelay(PG_SPIN_DEL);
+		else
+			pg_sleep(1);
+	}
+
+	to = time_after_eq(jiffies, tmo);
+
+	if ((r & (STAT_ERR & stop)) || to) {
+		s = read_reg(dev, 7);
+		e = read_reg(dev, 1);
+		p = read_reg(dev, 2);
+		if (verbose > 1)
+			printk("%s: %s: stat=0x%x err=0x%x phase=%d%s\n",
+			       dev->name, msg, s, e, p, to ? " timeout" : "");
+		if (to)
+			e |= 0x100;
+		dev->status = (e >> 4) & 0xff;
+		return -1;
+	}
+	return 0;
+}
+
+static int pg_command(struct pg *dev, char *cmd, int dlen, unsigned long tmo)
+{
+	int k;
+
+	pi_connect(dev->pi);
+
+	write_reg(dev, 6, DRIVE(dev));
+
+	if (pg_wait(dev, STAT_BUSY | STAT_DRQ, 0, tmo, "before command"))
+		goto fail;
+
+	write_reg(dev, 4, dlen % 256);
+	write_reg(dev, 5, dlen / 256);
+	write_reg(dev, 7, 0xa0);	/* ATAPI packet command */
+
+	if (pg_wait(dev, STAT_BUSY, STAT_DRQ, tmo, "command DRQ"))
+		goto fail;
+
+	if (read_reg(dev, 2) != 1) {
+		printk("%s: command phase error\n", dev->name);
+		goto fail;
+	}
+
+	pi_write_block(dev->pi, cmd, 12);
+
+	if (verbose > 1) {
+		printk("%s: Command sent, dlen=%d packet= ", dev->name, dlen);
+		for (k = 0; k < 12; k++)
+			printk("%02x ", cmd[k] & 0xff);
+		printk("\n");
+	}
+	return 0;
+fail:
+	pi_disconnect(dev->pi);
+	return -1;
+}
+
+static int pg_completion(struct pg *dev, char *buf, unsigned long tmo)
+{
+	int r, d, n, p;
+
+	r = pg_wait(dev, STAT_BUSY, STAT_DRQ | STAT_READY | STAT_ERR,
+		    tmo, "completion");
+
+	dev->dlen = 0;
+
+	while (read_reg(dev, 7) & STAT_DRQ) {
+		d = (read_reg(dev, 4) + 256 * read_reg(dev, 5));
+		n = ((d + 3) & 0xfffc);
+		p = read_reg(dev, 2) & 3;
+		if (p == 0)
+			pi_write_block(dev->pi, buf, n);
+		if (p == 2)
+			pi_read_block(dev->pi, buf, n);
+		if (verbose > 1)
+			printk("%s: %s %d bytes\n", dev->name,
+			       p ? "Read" : "Write", n);
+		dev->dlen += (1 - p) * d;
+		buf += d;
+		r = pg_wait(dev, STAT_BUSY, STAT_DRQ | STAT_READY | STAT_ERR,
+			    tmo, "completion");
+	}
+
+	pi_disconnect(dev->pi);
+
+	return r;
+}
+
+static int pg_reset(struct pg *dev)
+{
+	int i, k, err;
+	int expect[5] = { 1, 1, 1, 0x14, 0xeb };
+	int got[5];
+
+	pi_connect(dev->pi);
+	write_reg(dev, 6, DRIVE(dev));
+	write_reg(dev, 7, 8);
+
+	pg_sleep(20 * HZ / 1000);
+
+	k = 0;
+	while ((k++ < PG_RESET_TMO) && (status_reg(dev) & STAT_BUSY))
+		pg_sleep(1);
+
+	for (i = 0; i < 5; i++)
+		got[i] = read_reg(dev, i + 1);
+
+	err = memcmp(expect, got, sizeof(got)) ? -1 : 0;
+
+	if (verbose) {
+		printk("%s: Reset (%d) signature = ", dev->name, k);
+		for (i = 0; i < 5; i++)
+			printk("%3x", got[i]);
+		if (err)
+			printk(" (incorrect)");
+		printk("\n");
+	}
+
+	pi_disconnect(dev->pi);
+	return err;
+}
+
+static void xs(char *buf, char *targ, int len)
+{
+	char l = '\0';
+	int k;
+
+	for (k = 0; k < len; k++) {
+		char c = *buf++;
+		if (c != ' ' && c != l)
+			l = *targ++ = c;
+	}
+	if (l == ' ')
+		targ--;
+	*targ = '\0';
+}
+
+static int pg_identify(struct pg *dev, int log)
+{
+	int s;
+	char *ms[2] = { "master", "slave" };
+	char mf[10], id[18];
+	char id_cmd[12] = { ATAPI_IDENTIFY, 0, 0, 0, 36, 0, 0, 0, 0, 0, 0, 0 };
+	char buf[36];
+
+	s = pg_command(dev, id_cmd, 36, jiffies + PG_TMO);
+	if (s)
+		return -1;
+	s = pg_completion(dev, buf, jiffies + PG_TMO);
+	if (s)
+		return -1;
+
+	if (log) {
+		xs(buf + 8, mf, 8);
+		xs(buf + 16, id, 16);
+		printk("%s: %s %s, %s\n", dev->name, mf, id, ms[dev->drive]);
+	}
+
+	return 0;
+}
+
+/*
+ * returns  0, with id set if drive is detected
+ *	   -1, if drive detection failed
+ */
+static int pg_probe(struct pg *dev)
+{
+	if (dev->drive == -1) {
+		for (dev->drive = 0; dev->drive <= 1; dev->drive++)
+			if (!pg_reset(dev))
+				return pg_identify(dev, 1);
+	} else {
+		if (!pg_reset(dev))
+			return pg_identify(dev, 1);
+	}
+	return -1;
+}
+
+static int pg_detect(void)
+{
+	struct pg *dev = &devices[0];
+	int k, unit;
+
+	printk("%s: %s version %s, major %d\n", name, name, PG_VERSION, major);
+
+	par_drv = pi_register_driver(name);
+	if (!par_drv) {
+		pr_err("failed to register %s driver\n", name);
+		return -1;
+	}
+
+	k = 0;
+	if (pg_drive_count == 0) {
+		if (pi_init(dev->pi, 1, -1, -1, -1, -1, -1, pg_scratch,
+			    PI_PG, verbose, dev->name)) {
+			if (!pg_probe(dev)) {
+				dev->present = 1;
+				k++;
+			} else
+				pi_release(dev->pi);
+		}
+
+	} else
+		for (unit = 0; unit < PG_UNITS; unit++, dev++) {
+			int *parm = *drives[unit];
+			if (!parm[D_PRT])
+				continue;
+			if (pi_init(dev->pi, 0, parm[D_PRT], parm[D_MOD],
+				    parm[D_UNI], parm[D_PRO], parm[D_DLY],
+				    pg_scratch, PI_PG, verbose, dev->name)) {
+				if (!pg_probe(dev)) {
+					dev->present = 1;
+					k++;
+				} else
+					pi_release(dev->pi);
+			}
+		}
+
+	if (k)
+		return 0;
+
+	pi_unregister_driver(par_drv);
+	printk("%s: No ATAPI device detected\n", name);
+	return -1;
+}
+
+static int pg_open(struct inode *inode, struct file *file)
+{
+	int unit = iminor(inode) & 0x7f;
+	struct pg *dev = &devices[unit];
+	int ret = 0;
+
+	mutex_lock(&pg_mutex);
+	if ((unit >= PG_UNITS) || (!dev->present)) {
+		ret = -ENODEV;
+		goto out;
+	}
+
+	if (test_and_set_bit(0, &dev->access)) {
+		ret = -EBUSY;
+		goto out;
+	}
+
+	if (dev->busy) {
+		pg_reset(dev);
+		dev->busy = 0;
+	}
+
+	pg_identify(dev, (verbose > 1));
+
+	dev->bufptr = kmalloc(PG_MAX_DATA, GFP_KERNEL);
+	if (dev->bufptr == NULL) {
+		clear_bit(0, &dev->access);
+		printk("%s: buffer allocation failed\n", dev->name);
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	file->private_data = dev;
+
+out:
+	mutex_unlock(&pg_mutex);
+	return ret;
+}
+
+static int pg_release(struct inode *inode, struct file *file)
+{
+	struct pg *dev = file->private_data;
+
+	kfree(dev->bufptr);
+	dev->bufptr = NULL;
+	clear_bit(0, &dev->access);
+
+	return 0;
+}
+
+static ssize_t pg_write(struct file *filp, const char __user *buf, size_t count, loff_t *ppos)
+{
+	struct pg *dev = filp->private_data;
+	struct pg_write_hdr hdr;
+	int hs = sizeof (hdr);
+
+	if (dev->busy)
+		return -EBUSY;
+	if (count < hs)
+		return -EINVAL;
+
+	if (copy_from_user(&hdr, buf, hs))
+		return -EFAULT;
+
+	if (hdr.magic != PG_MAGIC)
+		return -EINVAL;
+	if (hdr.dlen < 0 || hdr.dlen > PG_MAX_DATA)
+		return -EINVAL;
+	if ((count - hs) > PG_MAX_DATA)
+		return -EINVAL;
+
+	if (hdr.func == PG_RESET) {
+		if (count != hs)
+			return -EINVAL;
+		if (pg_reset(dev))
+			return -EIO;
+		return count;
+	}
+
+	if (hdr.func != PG_COMMAND)
+		return -EINVAL;
+
+	dev->start = jiffies;
+	dev->timeout = hdr.timeout * HZ + HZ / 2 + jiffies;
+
+	if (pg_command(dev, hdr.packet, hdr.dlen, jiffies + PG_TMO)) {
+		if (dev->status & 0x10)
+			return -ETIME;
+		return -EIO;
+	}
+
+	dev->busy = 1;
+
+	if (copy_from_user(dev->bufptr, buf + hs, count - hs))
+		return -EFAULT;
+	return count;
+}
+
+static ssize_t pg_read(struct file *filp, char __user *buf, size_t count, loff_t *ppos)
+{
+	struct pg *dev = filp->private_data;
+	struct pg_read_hdr hdr;
+	int hs = sizeof (hdr);
+	int copy;
+
+	if (!dev->busy)
+		return -EINVAL;
+	if (count < hs)
+		return -EINVAL;
+
+	dev->busy = 0;
+
+	if (pg_completion(dev, dev->bufptr, dev->timeout))
+		if (dev->status & 0x10)
+			return -ETIME;
+
+	memset(&hdr, 0, sizeof(hdr));
+	hdr.magic = PG_MAGIC;
+	hdr.dlen = dev->dlen;
+	copy = 0;
+
+	if (hdr.dlen < 0) {
+		hdr.dlen = -1 * hdr.dlen;
+		copy = hdr.dlen;
+		if (copy > (count - hs))
+			copy = count - hs;
+	}
+
+	hdr.duration = (jiffies - dev->start + HZ / 2) / HZ;
+	hdr.scsi = dev->status & 0x0f;
+
+	if (copy_to_user(buf, &hdr, hs))
+		return -EFAULT;
+	if (copy > 0)
+		if (copy_to_user(buf + hs, dev->bufptr, copy))
+			return -EFAULT;
+	return copy + hs;
+}
+
+static int __init pg_init(void)
+{
+	int unit;
+	int err;
+
+	if (disable){
+		err = -EINVAL;
+		goto out;
+	}
+
+	pg_init_units();
+
+	if (pg_detect()) {
+		err = -ENODEV;
+		goto out;
+	}
+
+	err = register_chrdev(major, name, &pg_fops);
+	if (err < 0) {
+		printk("pg_init: unable to get major number %d\n", major);
+		for (unit = 0; unit < PG_UNITS; unit++) {
+			struct pg *dev = &devices[unit];
+			if (dev->present)
+				pi_release(dev->pi);
+		}
+		goto out;
+	}
+	major = err;	/* In case the user specified `major=0' (dynamic) */
+	pg_class = class_create(THIS_MODULE, "pg");
+	if (IS_ERR(pg_class)) {
+		err = PTR_ERR(pg_class);
+		goto out_chrdev;
+	}
+	for (unit = 0; unit < PG_UNITS; unit++) {
+		struct pg *dev = &devices[unit];
+		if (dev->present)
+			device_create(pg_class, NULL, MKDEV(major, unit), NULL,
+				      "pg%u", unit);
+	}
+	err = 0;
+	goto out;
+
+out_chrdev:
+	unregister_chrdev(major, "pg");
+out:
+	return err;
+}
+
+static void __exit pg_exit(void)
+{
+	int unit;
+
+	for (unit = 0; unit < PG_UNITS; unit++) {
+		struct pg *dev = &devices[unit];
+		if (dev->present)
+			device_destroy(pg_class, MKDEV(major, unit));
+	}
+	class_destroy(pg_class);
+	unregister_chrdev(major, name);
+
+	for (unit = 0; unit < PG_UNITS; unit++) {
+		struct pg *dev = &devices[unit];
+		if (dev->present)
+			pi_release(dev->pi);
+	}
+}
+
+MODULE_LICENSE("GPL");
+module_init(pg_init)
+module_exit(pg_exit)