Improve the performance for pipe(2) again. Also include some

fixes for previous version of new pipes from Bruce Evans.  This
new version:

Supports more properly the semantics of select (BDE).
Supports "OLD_PIPE" correctly (kern_descrip.c, BDE).
Eliminates incorrect EPIPE returns (bash 'pipe broken' messages.)
Much faster yet, currently tuned relatively conservatively -- but now
	gives approx 50% more perf than the new pipes code did originally.
	(That was about 50% more perf than the original BSD pipe code.)

Known bugs outstanding:
	No support for async io (SIGIO).  Will be included soon.

Next to do:
	Merge support for FIFOs.

Submitted by: bde

2 files changed, 449 insertions, 47 deletions

diff --git a/sys/kern/kern_descrip.c b/sys/kern/kern_descrip.c
index 5c781e5..d214f0d 100644
--- a/sys/kern/kern_descrip.c
+++ b/sys/kern/kern_descrip.c
@@ -36,7 +36,7 @@
  * SUCH DAMAGE.
  *
  *	@(#)kern_descrip.c	8.6 (Berkeley) 4/19/94
- * $Id: kern_descrip.c,v 1.23 1995/12/14 08:31:14 phk Exp $
+ * $Id: kern_descrip.c,v 1.24 1996/01/28 23:41:39 dyson Exp $
  */
 
 #include <sys/param.h>
@@ -425,9 +425,11 @@ ofstat(p, uap, retval)
 		error = soo_stat((struct socket *)fp->f_data, &ub);
 		break;
 
+#ifndef OLD_PIPE
 	case DTYPE_PIPE:
 		error = pipe_stat((struct pipe *)fp->f_data, &ub);
 		break;
+#endif
 
 	default:
 		panic("ofstat");
@@ -474,9 +476,11 @@ fstat(p, uap, retval)
 		error = soo_stat((struct socket *)fp->f_data, &ub);
 		break;
 
+#ifndef OLD_PIPE
 	case DTYPE_PIPE:
 		error = pipe_stat((struct pipe *)fp->f_data, &ub);
 		break;
+#endif
 
 	default:
 		panic("fstat");
@@ -512,7 +516,9 @@ fpathconf(p, uap, retval)
 		return (EBADF);
 	switch (fp->f_type) {
 
+#ifndef OLD_PIPE
 	case DTYPE_PIPE:
+#endif
 	case DTYPE_SOCKET:
 		if (uap->name != _PC_PIPE_BUF)
 			return (EINVAL);
diff --git a/sys/kern/sys_pipe.c b/sys/kern/sys_pipe.c
index dda604a..4e563c9 100644
--- a/sys/kern/sys_pipe.c
+++ b/sys/kern/sys_pipe.c
@@ -18,7 +18,7 @@
  * 5. Modifications may be freely made to this file if the above conditions
  *    are met.
  *
- * $Id: sys_pipe.c,v 1.3 1996/01/31 02:05:12 dyson Exp $
+ * $Id: sys_pipe.c,v 1.2 1996/01/29 02:57:33 dyson Exp $
  */
 
 #ifndef OLD_PIPE
@@ -30,6 +30,29 @@
  * do.
  */
 
+/*
+ * This code has two modes of operation, a small write mode and a large
+ * write mode.  The small write mode acts like conventional pipes with
+ * a kernel buffer.  If the buffer is less than PIPE_MINDIRECT, then the
+ * "normal" pipe buffering is done.  If the buffer is between PIPE_MINDIRECT
+ * and PIPE_SIZE in size, it is fully mapped and wired into the kernel, and
+ * the receiving process can copy it directly from the pages in the sending
+ * process.
+ *
+ * If the sending process receives a signal, it is possible that it will
+ * go away, and certainly it's address space can change, because control
+ * is returned back to the user-mode side.  In that case, the pipe code
+ * arranges to copy the buffer supplied by the user process, to a pageable
+ * kernel buffer, and the receiving process will grab the data from the
+ * pageable kernel buffer.  Since signals don't happen all that often,
+ * the copy operation is normally eliminated.
+ *
+ * The constant PIPE_MINDIRECT is chosen to make sure that buffering will
+ * happen for small transfers so that the system will not spend all of
+ * it's time context switching.  PIPE_SIZE is constrained by the
+ * amount of kernel virtual memory.
+ */
+
 #include <sys/param.h>
 #include <sys/systm.h>
 #include <sys/proc.h>
@@ -58,6 +81,7 @@
 #include <vm/vm_extern.h>
 #include <vm/pmap.h>
 #include <vm/vm_map.h>
+#include <vm/vm_page.h>
 
 static int pipe_read __P((struct file *fp, struct uio *uio, 
 		struct ucred *cred));
@@ -76,15 +100,34 @@ static struct fileops pipeops =
  * reference for performance reasons, so small amounts of outstanding I/O
  * will not wipe the cache.
  */
-#define PIPESIZE (16384)
-#define MINPIPESIZE (PIPESIZE/3)
-#define MAXPIPESIZE (2*PIPESIZE/3)
+#define MINPIPESIZE (PIPE_SIZE/3)
+#define MAXPIPESIZE (2*PIPE_SIZE/3)
+
+/*
+ * Maximum amount of kva for pipes -- this is kind-of a soft limit, but
+ * is there so that on large systems, we don't exhaust it.
+ */
+#define MAXPIPEKVA (8*1024*1024)
+
+/*
+ * Limit for direct transfers, we cannot, of course limit
+ * the amount of kva for pipes in general though.
+ */
+#define LIMITPIPEKVA (16*1024*1024)
+int amountpipekva;
 
 static void pipeclose __P((struct pipe *cpipe));
 static void pipebufferinit __P((struct pipe *cpipe));
 static void pipeinit __P((struct pipe *cpipe));
-static __inline int pipelock __P((struct pipe *cpipe));
+static __inline int pipelock __P((struct pipe *cpipe, int catch));
 static __inline void pipeunlock __P((struct pipe *cpipe));
+static int pipe_build_write_buffer __P((struct pipe *wpipe, struct uio *uio));
+static void pipe_destroy_write_buffer __P((struct pipe *wpipe));
+static int pipe_direct_write __P((struct pipe *wpipe, struct uio *uio));
+static void pipe_clone_write_buffer __P((struct pipe *wpipe));
+static void pipe_mark_pages_clean __P((struct pipe *cpipe));
+static int pipewrite __P((struct pipe *wpipe, struct uio *uio, int nbio));
+static void pipespace __P((struct pipe *cpipe));
 
 /*
  * The pipe system call for the DTYPE_PIPE type of pipes
@@ -106,8 +149,10 @@ pipe(p, uap, retval)
 
 	rpipe = malloc( sizeof (*rpipe), M_TEMP, M_WAITOK);
 	pipeinit(rpipe);
+	rpipe->pipe_state |= PIPE_DIRECTOK;
 	wpipe = malloc( sizeof (*wpipe), M_TEMP, M_WAITOK);
 	pipeinit(wpipe);
+	wpipe->pipe_state |= PIPE_DIRECTOK;
 
 	error = falloc(p, &rf, &fd);
 	if (error)
@@ -140,17 +185,13 @@ free1:
 	return (error);
 }
 
-/*
- * initialize and allocate VM and memory for pipe
- */
 static void
-pipeinit(cpipe)
+pipespace(cpipe)
 	struct pipe *cpipe;
 {
 	int npages, error;
 
-	npages = round_page(PIPESIZE)/PAGE_SIZE;
-
+	npages = round_page(cpipe->pipe_buffer.size)/PAGE_SIZE;
 	/*
 	 * Create an object, I don't like the idea of paging to/from
 	 * kernel_object.
@@ -163,16 +204,29 @@ pipeinit(cpipe)
 	 * The map entry is, by default, pageable.
 	 */
 	error = vm_map_find(kernel_map, cpipe->pipe_buffer.object, 0,
-		(vm_offset_t *) &cpipe->pipe_buffer.buffer, PIPESIZE, 1,
+		(vm_offset_t *) &cpipe->pipe_buffer.buffer, 
+		cpipe->pipe_buffer.size, 1,
 		VM_PROT_ALL, VM_PROT_ALL, 0);
 
 	if (error != KERN_SUCCESS)
 		panic("pipeinit: cannot allocate pipe -- out of kvm -- code = %d", error);
+	amountpipekva += cpipe->pipe_buffer.size;
+}
+
+/*
+ * initialize and allocate VM and memory for pipe
+ */
+static void
+pipeinit(cpipe)
+	struct pipe *cpipe;
+{
 
 	cpipe->pipe_buffer.in = 0;
 	cpipe->pipe_buffer.out = 0;
 	cpipe->pipe_buffer.cnt = 0;
-	cpipe->pipe_buffer.size = PIPESIZE;
+	cpipe->pipe_buffer.size = PIPE_SIZE;
+	/* Buffer kva gets dynamically allocated */
+	cpipe->pipe_buffer.buffer = NULL;
 
 	cpipe->pipe_state = 0;
 	cpipe->pipe_peer = NULL;
@@ -181,6 +235,14 @@ pipeinit(cpipe)
 	cpipe->pipe_atime = time;
 	cpipe->pipe_mtime = time;
 	bzero(&cpipe->pipe_sel, sizeof cpipe->pipe_sel);
+
+	/*
+	 * pipe data structure initializations to support direct pipe I/O
+	 */
+	cpipe->pipe_map.cnt = 0;
+	cpipe->pipe_map.kva = 0;
+	cpipe->pipe_map.pos = 0;
+	cpipe->pipe_map.npages = 0;
 }
 
 
@@ -188,13 +250,15 @@ pipeinit(cpipe)
  * lock a pipe for I/O, blocking other access
  */
 static __inline int
-pipelock(cpipe)
+pipelock(cpipe, catch)
 	struct pipe *cpipe;
+	int catch;
 {
 	int error;
 	while (cpipe->pipe_state & PIPE_LOCK) {
 		cpipe->pipe_state |= PIPE_LWANT;
-		if (error = tsleep( &cpipe->pipe_state, PRIBIO|PCATCH, "pipelk", 0)) {
+		if (error = tsleep( &cpipe->pipe_state,
+			catch?(PRIBIO|PCATCH):PRIBIO, "pipelk", 0)) {
 			return error;
 		}
 	}
@@ -217,6 +281,24 @@ pipeunlock(cpipe)
 	return;
 }
 
+#if 0
+static void
+pipe_mark_pages_clean(cpipe)
+	struct pipe *cpipe;
+{
+	vm_size_t off;
+	vm_page_t m;
+
+	for(off = 0; off < cpipe->pipe_buffer.object->size; off += 1) {
+		m = vm_page_lookup(cpipe->pipe_buffer.object, off);
+		if ((m != NULL) && (m->busy == 0) && (m->flags & PG_BUSY) == 0) {
+			m->dirty = 0;
+			pmap_clear_modify(VM_PAGE_TO_PHYS(m));
+		}
+	}
+}
+#endif
+
 /* ARGSUSED */
 static int
 pipe_read(fp, uio, cred)
@@ -228,16 +310,20 @@ pipe_read(fp, uio, cred)
 	struct pipe *rpipe = (struct pipe *) fp->f_data;
 	int error = 0;
 	int nread = 0;
+	int size;
 
 	++rpipe->pipe_busy;
 	while (uio->uio_resid) {
+		/*
+		 * normal pipe buffer receive
+		 */
 		if (rpipe->pipe_buffer.cnt > 0) {
 			int size = rpipe->pipe_buffer.size - rpipe->pipe_buffer.out;
 			if (size > rpipe->pipe_buffer.cnt)
 				size = rpipe->pipe_buffer.cnt;
 			if (size > uio->uio_resid)
 				size = uio->uio_resid;
-			if ((error = pipelock(rpipe)) == 0) {
+			if ((error = pipelock(rpipe,1)) == 0) {
 				error = uiomove( &rpipe->pipe_buffer.buffer[rpipe->pipe_buffer.out], 
 					size, uio);
 				pipeunlock(rpipe);
@@ -252,6 +338,29 @@ pipe_read(fp, uio, cred)
 			rpipe->pipe_buffer.cnt -= size;
 			nread += size;
 			rpipe->pipe_atime = time;
+		/*
+		 * Direct copy, bypassing a kernel buffer.
+		 */
+		} else if ((size = rpipe->pipe_map.cnt) &&
+			(rpipe->pipe_state & PIPE_DIRECTW)) {
+			caddr_t va;
+			if (size > uio->uio_resid)
+				size = uio->uio_resid;
+			if ((error = pipelock(rpipe,1)) == 0) {
+				va = (caddr_t) rpipe->pipe_map.kva + rpipe->pipe_map.pos;
+				error = uiomove(va, size, uio);
+				pipeunlock(rpipe);
+			}
+			if (error)
+				break;
+			nread += size;
+			rpipe->pipe_atime = time;
+			rpipe->pipe_map.pos += size;
+			rpipe->pipe_map.cnt -= size;
+			if (rpipe->pipe_map.cnt == 0) {
+				rpipe->pipe_state &= ~PIPE_DIRECTW;
+				wakeup(rpipe);
+			}
 		} else {
 			/*
 			 * detect EOF condition
@@ -272,8 +381,6 @@ pipe_read(fp, uio, cred)
 				error = EAGAIN;
 				break;
 			}
-			if (rpipe->pipe_peer == NULL)
-				break;
 
 			/*
 			 * If there is no more to read in the pipe, reset
@@ -281,7 +388,7 @@ pipe_read(fp, uio, cred)
 			 * cache hit stats.
 			 */
 		
-			if ((error = pipelock(rpipe)) == 0) {
+			if ((error = pipelock(rpipe,1)) == 0) {
 				if (rpipe->pipe_buffer.cnt == 0) {
 					rpipe->pipe_buffer.in = 0;
 					rpipe->pipe_buffer.out = 0;
@@ -307,8 +414,11 @@ pipe_read(fp, uio, cred)
 		 * it's pointers to the beginning.  This improves
 		 * cache hit stats.
 		 */
-		if ((error == 0) && (error = pipelock(rpipe)) == 0) {
+		if ((error == 0) && (error = pipelock(rpipe,1)) == 0) {
 			if (rpipe->pipe_buffer.cnt == 0) {
+#if 0
+				pipe_mark_pages_clean(rpipe);
+#endif
 				rpipe->pipe_buffer.in = 0;
 				rpipe->pipe_buffer.out = 0;
 			}
@@ -330,35 +440,273 @@ pipe_read(fp, uio, cred)
 	return error;
 }
 
-/* ARGSUSED */
+/*
+ * Map the sending processes' buffer into kernel space and wire it.
+ * This is similar to a physical write operation.
+ */
 static int
-pipe_write(fp, uio, cred)
-	struct file *fp;
+pipe_build_write_buffer(wpipe, uio)
+	struct pipe *wpipe;
 	struct uio *uio;
-	struct ucred *cred;
 {
-	struct pipe *rpipe = (struct pipe *) fp->f_data;
-	struct pipe *wpipe = rpipe->pipe_peer;
+	int size;
+	int i;
+	vm_offset_t addr, endaddr, paddr;
+
+	size = uio->uio_iov->iov_len;
+	if (size > wpipe->pipe_buffer.size)
+		size = wpipe->pipe_buffer.size;
+
+	endaddr = round_page(uio->uio_iov->iov_base + size);
+	for(i = 0, addr = trunc_page(uio->uio_iov->iov_base);
+		addr < endaddr;
+		addr += PAGE_SIZE, i+=1) {
+
+		vm_page_t m;
+
+		vm_fault_quick( addr, VM_PROT_READ);
+		paddr = pmap_kextract(addr);
+		if (!paddr) {
+			int j;
+			for(j=0;j<i;j++)
+				vm_page_unwire(wpipe->pipe_map.ms[j]);
+			return EFAULT;
+		}
+
+		m = PHYS_TO_VM_PAGE(paddr);
+		vm_page_wire(m);
+		wpipe->pipe_map.ms[i] = m;
+	}
+
+/*
+ * set up the control block
+ */
+	wpipe->pipe_map.npages = i;
+	wpipe->pipe_map.pos = ((vm_offset_t) uio->uio_iov->iov_base) & PAGE_MASK;
+	wpipe->pipe_map.cnt = size;
+
+/*
+ * and map the buffer
+ */
+	if (wpipe->pipe_map.kva == 0) {
+		wpipe->pipe_map.kva = kmem_alloc_pageable(kernel_map,
+			wpipe->pipe_buffer.size);
+		amountpipekva += wpipe->pipe_buffer.size;
+	}
+	pmap_qenter(wpipe->pipe_map.kva, wpipe->pipe_map.ms,
+		wpipe->pipe_map.npages);
+
+/*
+ * and update the uio data
+ */
+
+	uio->uio_iov->iov_len -= size;
+	uio->uio_iov->iov_base += size;
+	if (uio->uio_iov->iov_len == 0)
+		uio->uio_iov++;
+	uio->uio_resid -= size;
+	uio->uio_offset += size;
+	return 0;
+}
+
+/*
+ * unmap and unwire the process buffer
+ */
+static void
+pipe_destroy_write_buffer(wpipe)
+struct pipe *wpipe;
+{
+	int i;
+	pmap_qremove(wpipe->pipe_map.kva, wpipe->pipe_map.npages);
+
+	if (wpipe->pipe_map.kva) {
+		if (amountpipekva > MAXPIPEKVA) {
+			vm_offset_t kva = wpipe->pipe_map.kva;
+			wpipe->pipe_map.kva = 0;
+			kmem_free(kernel_map, kva,
+				wpipe->pipe_buffer.size);
+			amountpipekva -= wpipe->pipe_buffer.size;
+		}
+	}
+	for (i=0;i<wpipe->pipe_map.npages;i++)
+		vm_page_unwire(wpipe->pipe_map.ms[i]);
+}
+
+/*
+ * In the case of a signal, the writing process might go away.  This
+ * code copies the data into the circular buffer so that the source
+ * pages can be freed without loss of data.
+ */
+static void
+pipe_clone_write_buffer(wpipe)
+struct pipe *wpipe;
+{
+	int size;
+	int pos;
+
+	size = wpipe->pipe_map.cnt;
+	pos = wpipe->pipe_map.pos;
+	bcopy((caddr_t) wpipe->pipe_map.kva+pos,
+			(caddr_t) wpipe->pipe_buffer.buffer,
+			size);
+
+	wpipe->pipe_buffer.in = size;
+	wpipe->pipe_buffer.out = 0;
+	wpipe->pipe_buffer.cnt = size;
+	wpipe->pipe_state &= ~PIPE_DIRECTW;
+
+	pipe_destroy_write_buffer(wpipe);
+}
+
+/*
+ * This implements the pipe buffer write mechanism.  Note that only
+ * a direct write OR a normal pipe write can be pending at any given time.
+ * If there are any characters in the pipe buffer, the direct write will
+ * be deferred until the receiving process grabs all of the bytes from
+ * the pipe buffer.  Then the direct mapping write is set-up.
+ */
+static int
+pipe_direct_write(wpipe, uio)
+	struct pipe *wpipe;
+	struct uio *uio;
+{
+	int error;
+	while (wpipe->pipe_state & PIPE_DIRECTW) {
+		error = tsleep(wpipe,
+				PRIBIO|PCATCH, "pipdww", 0);
+		if (error || (wpipe->pipe_state & PIPE_EOF))
+			goto error1;
+	}
+	wpipe->pipe_map.cnt = 0;	/* transfer not ready yet */
+	wpipe->pipe_state |= PIPE_DIRECTW;
+	while (wpipe->pipe_buffer.cnt > 0) {
+		error = tsleep(wpipe,
+				PRIBIO|PCATCH, "pipdwc", 0);
+		if (error || (wpipe->pipe_state & PIPE_EOF)) {
+			wpipe->pipe_state &= ~PIPE_DIRECTW;
+			if (error == 0)
+				error = EPIPE;
+			goto error1;
+		}
+	}
+
+	error = pipe_build_write_buffer(wpipe, uio);
+	if (error) {
+		wpipe->pipe_state &= ~PIPE_DIRECTW;
+		goto error1;
+	}
+
+	if (wpipe->pipe_state & PIPE_WANTR) {
+		wpipe->pipe_state &= ~PIPE_WANTR;
+		wakeup(wpipe);
+	}
+
+	error = 0;
+	while (!error && (wpipe->pipe_state & PIPE_DIRECTW)) {
+		if (wpipe->pipe_state & PIPE_EOF) {
+			pipelock(wpipe, 0);
+			pipe_destroy_write_buffer(wpipe);
+			pipeunlock(wpipe);
+			wakeup(wpipe);
+			return EPIPE;
+		}
+		error = tsleep(wpipe, PRIBIO|PCATCH, "pipdwt", 0);
+	}
+
+	pipelock(wpipe,0);
+	if (wpipe->pipe_state & PIPE_DIRECTW) {
+		/*
+		 * this bit of trickery substitutes a kernel buffer for
+		 * the process that might be going away.
+		 */
+		pipe_clone_write_buffer(wpipe);
+	} else {
+		pipe_destroy_write_buffer(wpipe);
+	}
+	pipeunlock(wpipe);
+	return error;
+
+error1:
+	wakeup(wpipe);
+	return error;
+}
+	
+static __inline int
+pipewrite(wpipe, uio, nbio)
+	struct pipe *wpipe;
+	struct uio *uio;
+	int nbio;
+{
 	int error = 0;
 
 	/*
 	 * detect loss of pipe read side, issue SIGPIPE if lost.
 	 */
 	if (wpipe == NULL || (wpipe->pipe_state & PIPE_EOF)) {
-		psignal(curproc, SIGPIPE);
 		return EPIPE;
 	}
 
+	if( wpipe->pipe_buffer.buffer == NULL) {
+		if ((error = pipelock(wpipe,1)) == 0) {
+			pipespace(wpipe);
+			pipeunlock(wpipe);
+		} else {
+			return error;
+		}
+	}
+
 	++wpipe->pipe_busy;
 	while (uio->uio_resid) {
-		int space = wpipe->pipe_buffer.size - wpipe->pipe_buffer.cnt;
-		if (space > 0) {
+		int space;
+		/*
+		 * If the transfer is large, we can gain performance if
+		 * we do process-to-process copies directly.
+		 */
+		if ((amountpipekva < LIMITPIPEKVA) &&
+			(uio->uio_iov->iov_len >= PIPE_MINDIRECT)) {
+			error = pipe_direct_write( wpipe, uio);
+			if (error) {
+				break;
+			}
+			continue;
+		}
+
+		/*
+		 * Pipe buffered writes cannot be coincidental with
+		 * direct writes.  We wait until the currently executing
+		 * direct write is completed before we start filling the
+		 * pipe buffer.
+		 */
+	retrywrite:
+		while (wpipe->pipe_state & PIPE_DIRECTW) {
+			error = tsleep(wpipe,
+					PRIBIO|PCATCH, "pipbww", 0);
+			if (error)
+				break;
+		}
+
+		space = wpipe->pipe_buffer.size - wpipe->pipe_buffer.cnt;
+
+		/*
+		 * We must afford contiguous writes on buffers of size
+		 * PIPE_BUF or less.
+		 */
+		if ((space > 0) &&
+			((uio->uio_resid > PIPE_BUF) || (uio->uio_resid <= space))) {
 			int size = wpipe->pipe_buffer.size - wpipe->pipe_buffer.in;
 			if (size > space)
 				size = space;
 			if (size > uio->uio_resid)
 				size = uio->uio_resid;
-			if ((error = pipelock(wpipe)) == 0) {
+			if ((error = pipelock(wpipe,1)) == 0) {
+				/*
+				 * It is possible for a direct write to
+				 * slip in on us... handle it here...
+				 */
+				if (wpipe->pipe_state & PIPE_DIRECTW) {
+					pipeunlock(wpipe);
+					goto retrywrite;
+				}
 				error = uiomove( &wpipe->pipe_buffer.buffer[wpipe->pipe_buffer.in], 
 					size, uio);
 				pipeunlock(wpipe);
@@ -383,9 +731,11 @@ pipe_write(fp, uio, cred)
 			/*
 			 * don't block on non-blocking I/O
 			 */
-			if (wpipe->pipe_state & PIPE_NBIO) {
+			if (nbio) {
+				error = EAGAIN;
 				break;
 			}
+
 			wpipe->pipe_state |= PIPE_WANTW;
 			if (error = tsleep(wpipe, (PRIBIO+1)|PCATCH, "pipewr", 0)) {
 				break;
@@ -395,13 +745,12 @@ pipe_write(fp, uio, cred)
 			 * to ourselves.
 			 */
 			if (wpipe->pipe_state & PIPE_EOF) {
-				psignal(curproc, SIGPIPE);
 				error = EPIPE;
+				break;
 			}	
 		}
 	}
 
-	--wpipe->pipe_busy;
 	if ((wpipe->pipe_busy == 0) &&
 		(wpipe->pipe_state & PIPE_WANT)) {
 		wpipe->pipe_state &= ~(PIPE_WANT|PIPE_WANTR);
@@ -416,13 +765,32 @@ pipe_write(fp, uio, cred)
 			wakeup(wpipe);
 		}
 	}
+	if ((wpipe->pipe_buffer.cnt == 0) &&
+		(uio->uio_resid == 0) &&
+		(error == EPIPE))
+		error = 0;
+		
 	if (wpipe->pipe_state & PIPE_SEL) {
 		wpipe->pipe_state &= ~PIPE_SEL;
 		selwakeup(&wpipe->pipe_sel);
 	}
+
+	--wpipe->pipe_busy;
 	return error;
 }
 
+/* ARGSUSED */
+static int
+pipe_write(fp, uio, cred)
+	struct file *fp;
+	struct uio *uio;
+	struct ucred *cred;
+{
+	struct pipe *rpipe = (struct pipe *) fp->f_data;
+	struct pipe *wpipe = rpipe->pipe_peer;
+	return pipewrite(wpipe, uio, (rpipe->pipe_state & PIPE_NBIO)?1:0);
+}
+
 /*
  * we implement a very minimal set of ioctls for compatibility with sockets.
  */
@@ -482,7 +850,8 @@ pipe_select(fp, which, p)
 	switch (which) {
 
 	case FREAD:
-		if (rpipe->pipe_buffer.cnt > 0) {
+		if (rpipe->pipe_buffer.cnt > 0 ||
+			(rpipe->pipe_state & PIPE_EOF)) {
 			splx(s);
 			return (1);
 		}
@@ -491,11 +860,9 @@ pipe_select(fp, which, p)
 		break;
 
 	case FWRITE:
-		if (wpipe == 0) {
-			splx(s);
-			return (1);
-		}
-		if (wpipe->pipe_buffer.cnt < wpipe->pipe_buffer.size) {
+		if ((wpipe == NULL) ||
+			(wpipe->pipe_state & PIPE_EOF) ||
+			((wpipe->pipe_buffer.size - wpipe->pipe_buffer.cnt) >= PIPE_BUF)) {
 			splx(s);
 			return (1);
 		}
@@ -504,6 +871,13 @@ pipe_select(fp, which, p)
 		break;
 
 	case 0:
+		if ((rpipe->pipe_state & PIPE_EOF) ||
+			(wpipe == NULL) ||
+			(wpipe->pipe_state & PIPE_EOF)) {
+			splx(s);
+			return (1);
+		}
+			
 		selrecord(p, &rpipe->pipe_sel);
 		rpipe->pipe_state |= PIPE_SEL;
 		break;
@@ -519,7 +893,7 @@ pipe_stat(pipe, ub)
 {
 	bzero((caddr_t)ub, sizeof (*ub));
 	ub->st_mode = S_IFSOCK;
-	ub->st_blksize = pipe->pipe_buffer.size / 2;
+	ub->st_blksize = pipe->pipe_buffer.size;
 	ub->st_size = pipe->pipe_buffer.cnt;
 	ub->st_blocks = (ub->st_size + ub->st_blksize - 1) / ub->st_blksize;
 	TIMEVAL_TO_TIMESPEC(&pipe->pipe_atime, &ub->st_atimespec);
@@ -548,7 +922,14 @@ static void
 pipeclose(cpipe)
 	struct pipe *cpipe;
 {
+	struct pipe *ppipe;
 	if (cpipe) {
+		
+		if (cpipe->pipe_state & PIPE_SEL) {
+			cpipe->pipe_state &= ~PIPE_SEL;
+			selwakeup(&cpipe->pipe_sel);
+		}
+
 		/*
 		 * If the other side is blocked, wake it up saying that
 		 * we want to close it down.
@@ -562,17 +943,32 @@ pipeclose(cpipe)
 		/*
 		 * Disconnect from peer
 		 */
-		if (cpipe->pipe_peer) {
-			cpipe->pipe_peer->pipe_state |= PIPE_EOF;
-			wakeup(cpipe->pipe_peer);
-			cpipe->pipe_peer->pipe_peer = NULL;
+		if (ppipe = cpipe->pipe_peer) {
+			if (ppipe->pipe_state & PIPE_SEL) {
+				ppipe->pipe_state &= ~PIPE_SEL;
+				selwakeup(&ppipe->pipe_sel);
+			}
+
+			ppipe->pipe_state |= PIPE_EOF;
+			wakeup(ppipe);
+			ppipe->pipe_peer = NULL;
 		}
 
 		/*
 		 * free resources
 		 */
-		kmem_free(kernel_map, (vm_offset_t)cpipe->pipe_buffer.buffer,
-			cpipe->pipe_buffer.size);
+		if (cpipe->pipe_buffer.buffer) {
+			amountpipekva -= cpipe->pipe_buffer.size;
+			kmem_free(kernel_map,
+				(vm_offset_t)cpipe->pipe_buffer.buffer,
+				cpipe->pipe_buffer.size);
+		}
+		if (cpipe->pipe_map.kva) {
+			amountpipekva -= cpipe->pipe_buffer.size;
+			kmem_free(kernel_map,
+				cpipe->pipe_map.kva,
+				cpipe->pipe_buffer.size);
+		}
 		free(cpipe, M_TEMP);
 	}
 }