linux/drivers/block/aoe/aoe.h
Ed L. Cashin 9bb237b6a6 aoe: dynamically allocate a capped number of skbs when necessary
What this Patch Does

  Even before this recent series of 12 patches to 2.6.22-rc4, the aoe
  driver was reusing a small set of skbs that were allocated once and
  were only used for outbound AoE commands.

  The network layer cannot be allowed to put_page on the data that is
  still associated with a bio we haven't returned to the block layer,
  so the aoe driver (even before the patch under discussion) is still
  the owner of skbs that have been handed to the network layer for
  transmission.  We need to keep track of these skbs so that we can
  free them, but by tracking them, we can also easily re-use them.

  The new patch was a response to the behavior of certain network
  drivers.  We cannot reuse an skb that the network driver still has
  in its transmit ring.  Network drivers can defer transmit ring
  cleanup and then use the state in the skb to determine how many data
  segments to clean up in its transmit ring.  The tg3 driver is one
  driver that behaves in this way.

  When the network driver defers cleanup of its transmit ring, the aoe
  driver can find itself in a situation where it would like to send an
  AoE command, and the AoE target is ready for more work, but the
  network driver still has all of the pre-allocated skbs.  In that
  case, the new patch just calls alloc_skb, as you'd expect.

  We don't want to get carried away, though.  We try not to do
  excessive allocation in the write path, so we cap the number of skbs
  we dynamically allocate.

  Probably calling it a "dynamic pool" is misleading.  We were already
  trying to use a small fixed-size set of pre-allocated skbs before
  this patch, and this patch just provides a little headroom (with a
  ceiling, though) to accomodate network drivers that hang onto skbs,
  by allocating when needed.  The d->skbpool_hd list of allocated skbs
  is necessary so that we can free them later.

  We didn't notice the need for this headroom until AoE targets got
  fast enough.

Alternatives

  If the network layer never did a put_page on the pages in the bio's
  we get from the block layer, then it would be possible for us to
  hand skbs to the network layer and forget about them, allowing the
  network layer to free skbs itself (and thereby calling our own
  skb->destructor callback function if we needed that).  In that case
  we could get rid of the pre-allocated skbs and also the
  d->skbpool_hd, instead just calling alloc_skb every time we wanted
  to transmit a packet.  The slab allocator would effectively maintain
  the list of skbs.

  Besides a loss of CPU cache locality, the main concern with that
  approach the danger that it would increase the likelihood of
  deadlock when VM is trying to free pages by writing dirty data from
  the page cache through the aoe driver out to persistent storage on
  an AoE device.  Right now we have a situation where we have
  pre-allocation that corresponds to how much we use, which seems
  ideal.

  Of course, there's still the separate issue of receiving the packets
  that tell us that a write has successfully completed on the AoE
  target.  When memory is low and VM is using AoE to flush dirty data
  to free up pages, it would be perfect if there were a way for us to
  register a fast callback that could recognize write command
  completion responses.  But I don't think the current problems with
  the receive side of the situation are a justification for
  exacerbating the problem on the transmit side.

Signed-off-by: Ed L. Cashin <ecashin@coraid.com>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-02-08 09:22:32 -08:00

208 lines
4.8 KiB
C

/* Copyright (c) 2006 Coraid, Inc. See COPYING for GPL terms. */
#define VERSION "47"
#define AOE_MAJOR 152
#define DEVICE_NAME "aoe"
/* set AOE_PARTITIONS to 1 to use whole-disks only
* default is 16, which is 15 partitions plus the whole disk
*/
#ifndef AOE_PARTITIONS
#define AOE_PARTITIONS (16)
#endif
#define SYSMINOR(aoemajor, aoeminor) ((aoemajor) * NPERSHELF + (aoeminor))
#define AOEMAJOR(sysminor) ((sysminor) / NPERSHELF)
#define AOEMINOR(sysminor) ((sysminor) % NPERSHELF)
#define WHITESPACE " \t\v\f\n"
enum {
AOECMD_ATA,
AOECMD_CFG,
AOEFL_RSP = (1<<3),
AOEFL_ERR = (1<<2),
AOEAFL_EXT = (1<<6),
AOEAFL_DEV = (1<<4),
AOEAFL_ASYNC = (1<<1),
AOEAFL_WRITE = (1<<0),
AOECCMD_READ = 0,
AOECCMD_TEST,
AOECCMD_PTEST,
AOECCMD_SET,
AOECCMD_FSET,
AOE_HVER = 0x10,
};
struct aoe_hdr {
unsigned char dst[6];
unsigned char src[6];
__be16 type;
unsigned char verfl;
unsigned char err;
__be16 major;
unsigned char minor;
unsigned char cmd;
__be32 tag;
};
struct aoe_atahdr {
unsigned char aflags;
unsigned char errfeat;
unsigned char scnt;
unsigned char cmdstat;
unsigned char lba0;
unsigned char lba1;
unsigned char lba2;
unsigned char lba3;
unsigned char lba4;
unsigned char lba5;
unsigned char res[2];
};
struct aoe_cfghdr {
__be16 bufcnt;
__be16 fwver;
unsigned char scnt;
unsigned char aoeccmd;
unsigned char cslen[2];
};
enum {
DEVFL_UP = 1, /* device is installed in system and ready for AoE->ATA commands */
DEVFL_TKILL = (1<<1), /* flag for timer to know when to kill self */
DEVFL_EXT = (1<<2), /* device accepts lba48 commands */
DEVFL_CLOSEWAIT = (1<<3), /* device is waiting for all closes to revalidate */
DEVFL_GDALLOC = (1<<4), /* need to alloc gendisk */
DEVFL_KICKME = (1<<5), /* slow polling network card catch */
DEVFL_NEWSIZE = (1<<6), /* need to update dev size in block layer */
BUFFL_FAIL = 1,
};
enum {
DEFAULTBCNT = 2 * 512, /* 2 sectors */
NPERSHELF = 16, /* number of slots per shelf address */
FREETAG = -1,
MIN_BUFS = 16,
NTARGETS = 8,
NAOEIFS = 8,
NSKBPOOLMAX = 128,
TIMERTICK = HZ / 10,
MINTIMER = HZ >> 2,
MAXTIMER = HZ << 1,
HELPWAIT = 20,
};
struct buf {
struct list_head bufs;
ulong stime; /* for disk stats */
ulong flags;
ulong nframesout;
ulong resid;
ulong bv_resid;
ulong bv_off;
sector_t sector;
struct bio *bio;
struct bio_vec *bv;
};
struct frame {
int tag;
ulong waited;
struct buf *buf;
char *bufaddr;
ulong bcnt;
sector_t lba;
struct sk_buff *skb;
};
struct aoeif {
struct net_device *nd;
unsigned char lost;
unsigned char lostjumbo;
ushort maxbcnt;
};
struct aoetgt {
unsigned char addr[6];
ushort nframes;
struct frame *frames;
struct aoeif ifs[NAOEIFS];
struct aoeif *ifp; /* current aoeif in use */
ushort nout;
ushort maxout;
u16 lasttag; /* last tag sent */
u16 useme;
ulong lastwadj; /* last window adjustment */
int wpkts, rpkts;
int dataref;
};
struct aoedev {
struct aoedev *next;
ulong sysminor;
ulong aoemajor;
u16 aoeminor;
u16 flags;
u16 nopen; /* (bd_openers isn't available without sleeping) */
u16 rttavg; /* round trip average of requests/responses */
u16 mintimer;
u16 fw_ver; /* version of blade's firmware */
struct work_struct work;/* disk create work struct */
struct gendisk *gd;
struct request_queue blkq;
struct hd_geometry geo;
sector_t ssize;
struct timer_list timer;
spinlock_t lock;
struct sk_buff *sendq_hd; /* packets needing to be sent, list head */
struct sk_buff *sendq_tl;
struct sk_buff *skbpool_hd;
struct sk_buff *skbpool_tl;
int nskbpool;
mempool_t *bufpool; /* for deadlock-free Buf allocation */
struct list_head bufq; /* queue of bios to work on */
struct buf *inprocess; /* the one we're currently working on */
struct aoetgt *targets[NTARGETS];
struct aoetgt **tgt; /* target in use when working */
struct aoetgt **htgt; /* target needing rexmit assistance */
};
int aoeblk_init(void);
void aoeblk_exit(void);
void aoeblk_gdalloc(void *);
void aoedisk_rm_sysfs(struct aoedev *d);
int aoechr_init(void);
void aoechr_exit(void);
void aoechr_error(char *);
void aoecmd_work(struct aoedev *d);
void aoecmd_cfg(ushort aoemajor, unsigned char aoeminor);
void aoecmd_ata_rsp(struct sk_buff *);
void aoecmd_cfg_rsp(struct sk_buff *);
void aoecmd_sleepwork(struct work_struct *);
void aoecmd_cleanslate(struct aoedev *);
struct sk_buff *aoecmd_ata_id(struct aoedev *);
int aoedev_init(void);
void aoedev_exit(void);
struct aoedev *aoedev_by_aoeaddr(int maj, int min);
struct aoedev *aoedev_by_sysminor_m(ulong sysminor);
void aoedev_downdev(struct aoedev *d);
int aoedev_isbusy(struct aoedev *d);
int aoedev_flush(const char __user *str, size_t size);
int aoenet_init(void);
void aoenet_exit(void);
void aoenet_xmit(struct sk_buff *);
int is_aoe_netif(struct net_device *ifp);
int set_aoe_iflist(const char __user *str, size_t size);
unsigned long long mac_addr(char addr[6]);