SchOS/kernel/dev/disk.c

// Raw block devices: /dev/hda, /dev/hdb.
//
// Until this existed, nothing outside the kernel could touch a disk. sys_mount
// does not even look at the device it is given -- it mounts the one SchFS
// volume the kernel found at boot and ignores `source` -- so a program had no
// way to read a partition table, format a volume, or write a boot sector. An
// installer is exactly a program that needs all three.
//
// This is a character device in the devsw table rather than a separate block
// layer, because that is where the offset already is: fileread/filewrite pass
// f->off and advance it by whatever comes back, which is all a seekable device
// needs. `dd if=/dev/hdb bs=4096 count=1` works, and so does lseek.
//
// The minor number is the IDE device number: 0 is the disk the kernel booted
// from, 1 is the second channel. Same numbering iderw() uses, deliberately --
// two numbering schemes for the same disks is a bug waiting to be written.
//
// WARNING, and the reason this file is short: writes here bypass the
// filesystem entirely. Writing to a device with a mounted filesystem on it will
// corrupt that filesystem, because the buffer cache above still believes its
// own copy. That is inherent to a raw device and not a defect -- but it is why
// an installer must target a disk it has not mounted.

#include "core/types.h"
#include "core/defs.h"
#include "core/param.h"
#include "core/spinlock.h"
#include "core/sleeplock.h"
#include "fs/fs.h"
#include "fs/buf.h"
#include "fs/file.h"
#include "uapi/errno.h"
#include "fs/blkdev.h"

// The byte offset is 64-bit, all the way from struct file's off through the
// devsw signature to bread's block number. It used to narrow to a 32-bit uint
// at the devsw boundary, which capped a raw device at 4 GB -- and the constant
// that documented the cap was never referenced by anything, so the failure was
// not a refusal at 4 GB but silent wraparound past it. Nothing here may narrow
// it again: an NVMe namespace or any modern disk is far beyond that.

// A disk or partition that registered with the block layer (minor 4 onward):
// a SATA or NVMe disk, a USB stick, any partition on them. Moved through
// blkdev_xfer in whole sectors rather than through the buffer cache, because a
// partition need not be a whole number of BSIZE blocks and the buffer cache's
// last block of one would run into the next. A read at the end returns 0, as
// a file's does, and a write there fails with ENOSPC.
static int
blk_rw(uint dev, uint64 off, char *p, int n, int write)
{
  uint32 ss = 0;
  uint64 nsec = blkdev_sectors(dev, &ss);
  if (nsec == 0)
    return -ENXIO;
  if (n <= 0)
    return 0;
  uint64 size = nsec * ss;
  if (off >= size)
    return write ? -ENOSPC : 0;
  if ((uint64)n > size - off)
    n = (int)(size - off);

  char *tmp = kalloc();
  if (!tmp)
    return -ENOMEM;
  int done = 0;
  while (done < n) {
    uint64 pos = off + (uint64)done;
    uint64 lba = pos / ss;
    uint   in  = (uint)(pos % ss);
    uint32 cnt = (uint32)(BSIZE / ss);
    if (lba + cnt > nsec)
      cnt = (uint32)(nsec - lba);
    int want = (int)(cnt * ss) - (int)in;
    if (want > n - done)
      want = n - done;
    // Whole sectors are written as they are; a partial one is read first.
    uint32 rcnt = (in + want + ss - 1) / ss;
    if (!write || in || want % ss)
      if (blkdev_xfer(dev, lba, rcnt, tmp, 0) < 0)
        break;
    if (write) {
      memmove(tmp + in, p + done, want);
      if (blkdev_xfer(dev, lba, rcnt, tmp, 1) < 0)
        break;
    } else
      memmove(p + done, tmp + in, want);
    done += want;
  }
  kfree(tmp);
  return done ? done : -EIO;
}

int
diskread(struct inode *ip, uint64 off, char *dst, int n)
{
  int dev = ip->minor;

  if (dev >= BLKDEV_FIRST)
    return blk_rw((uint)dev, off, dst, n, 0);
  if (!ide_disk_present(dev))
    return -ENXIO;
  if (n <= 0)
    return 0;

  int done = 0;
  while (done < n) {
    uint64 pos   = off + (uint64)done;
    uint64 blk   = pos / BSIZE;
    uint   inblk = (uint)(pos % BSIZE);
    int    want  = BSIZE - (int)inblk;
    if (want > n - done)
      want = n - done;

    struct buf *b = bread((uint)dev, blk);
    memmove(dst + done, b->data + inblk, want);
    brelse(b);

    done += want;
  }
  return done;
}

int
diskwrite(struct inode *ip, uint64 off, char *src, int n)
{
  int dev = ip->minor;

  if (dev >= BLKDEV_FIRST)
    return blk_rw((uint)dev, off, src, n, 1);
  if (!ide_disk_present(dev))
    return -ENXIO;
  if (n <= 0)
    return 0;

  int done = 0;
  while (done < n) {
    uint64 pos   = off + (uint64)done;
    uint64 blk   = pos / BSIZE;
    uint   inblk = (uint)(pos % BSIZE);
    int    want  = BSIZE - (int)inblk;
    if (want > n - done)
      want = n - done;

    // Read-modify-write for anything that is not a whole block. bread() first
    // even when the whole block is being replaced would be a wasted revolution,
    // but distinguishing the two cases means trusting the caller's alignment;
    // the disk is not the bottleneck in an installer and correctness here is
    // worth more than the read.
    struct buf *b = bread((uint)dev, blk);
    memmove(b->data + inblk, src + done, want);
    bwrite(b);
    brelse(b);

    done += want;
  }
  return done;
}