search for: file_can_trim

...c @@ -1,5 +1,5 @@ /* nbdkit - * Copyright (C) 2013 Red Hat Inc. + * Copyright (C) 2013-2018 Red Hat Inc. * All rights reserved. * * Redistribution and use in source and binary forms, with or without @@ -175,6 +175,18 @@ file_get_size (void *handle) return statbuf.st_size; } +static int +file_can_trim (void *handle) +{ + /* Trim is advisory, but we prefer to advertise it only when we can + * actually (attempt to) punch holes. */ +#ifdef FALLOC_FL_PUNCH_HOLE + return 1; +#else + return 0; +#endif +} + /* Read data from the file. */ static int file_pread (void *handle, void *buf, uint32_t...

...s defined before we did the first call. Once we tried > + * to punch a hole, report the actual cappability of the underlying file. */ s/cappability/capability/ If you use a tri-state, then report true if the variable is indeterminate or works, false if known to fail. > static int > file_can_trim (void *handle) > { > - /* Trim is advisory, but we prefer to advertise it only when we can > - * actually (attempt to) punch holes. Since not all filesystems > - * support all fallocate modes, it would be nice if we had a way > - * from fpathconf() to definitively learn wha...

...ile/file.c index dcff0ee..628f8fb 100644 --- a/plugins/file/file.c +++ b/plugins/file/file.c @@ -250,6 +250,13 @@ file_get_size (void *handle) } } +/* Allow multiple parallel connections from a single client. */ +static int +file_can_multi_conn (void *handle) +{ + return 1; +} + static int file_can_trim (void *handle) { @@ -507,6 +514,7 @@ static struct nbdkit_plugin plugin = { .open = file_open, .close = file_close, .get_size = file_get_size, + .can_multi_conn = file_can_multi_conn, .can_trim = file_can_trim, .can_fua = file...

...ile/file.c index 2a8c9b5..dcff0ee 100644 --- a/plugins/file/file.c +++ b/plugins/file/file.c @@ -52,6 +52,8 @@ #include <linux/fs.h> /* For BLKZEROOUT */ #endif +#define NBDKIT_API_VERSION 2 + #include <nbdkit-plugin.h> #include "isaligned.h" @@ -263,9 +265,30 @@ file_can_trim (void *handle) #endif } +static int +file_can_fua (void *handle) +{ + return NBDKIT_FUA_NATIVE; +} + +/* Flush the file to disk. */ +static int +file_flush (void *handle, uint32_t flags) +{ + struct handle *h = handle; + + if (fdatasync (h->fd) == -1) { + nbdkit_error ("fdatasync:...

...> assumes it works). So the best you can do is make subsequent > connections more efficient after the initial connection has learned state. > Sure disabling trim does not help the client. The only advantage is not calling trim when it does nothing. Do you think it is better to leave the file_can_trim as is, to avoid confusion? > > > - zero changed to: > > 1. If we can punch hole and may trim, try PUNCH_HOLE > > 2. If we can zero range, try ZERO_RANGE > > 3. Fall back to manual writing > > > > - trim changed to: > > 1. If we can punch hol...

This is the third version to support efficient zero for block devices on older kernels (e.g. RHEL 7.5), and file systems that do not support yet FALLOC_FS_ZERO_RANGE (e.g. NFS 4.2). Changes since v2: - Revert file_can_trim change, since it is too late to change the value after negotiation. Changing the capability dinamically may be useful internally, but it should be done via other means. - Do not depend on FALLOC_FL_* when including <linux/fs.h>. - Add common/includes/isaligned.h for is_aligned helper, imp...

First thing to say is that I need to do a *lot* more testing on this, so this is just an early peek. In particular, although it passed ‘make check && make check-valgrind’ I have *not* tested it against a multi-conn-aware client such as the Linux kernel >= 4.9. This implements NBD_FLAG_CAN_MULTI_CONN, described in the protocol doc as: "NBD_FLAG_CAN_MULTI_CONN: Indicates that

For existing commits, this is almost identical to v1, except that I updated some commit messages and reordered the commits in a somewhat more logical sequence. The main changes are the extra commits: [06/11] plugins: Return NBD_FLAG_CAN_MULTI_CONN from some readonly plugins. - Readonly plugins that can set the flag unconditionally. [09/11] partitioning: Return NBD_FLAG_CAN_MULTI_CONN. [10/11]

...e prefer to advertise it only when we can actually + * (attempt to) punch holes. Before we tried to punch a hole, report true if + * FALLOC_FL_PUNCH_HOLE is defined before we did the first call. Once we tried + * to punch a hole, report the actual cappability of the underlying file. */ static int file_can_trim (void *handle) { - /* Trim is advisory, but we prefer to advertise it only when we can - * actually (attempt to) punch holes. Since not all filesystems - * support all fallocate modes, it would be nice if we had a way - * from fpathconf() to definitively learn what will work on a given -...

This is the second version to support efficient zero for block devices on older kernels (e.g. RHEL 7.5), and file systems that do not support yet FALLOC_FS_ZERO_RANGE (e.g. NFS 4.2). Changes since v1: - Split to smaller patches - Skip linux only includes on other systems - Skip code using BLKZEROOUT if the macro is not defined - Try BLKZEROOUT only if the offset and count are aligned to device

...t_error ("fstat: %m"); - return -1; - } - - return statbuf.st_size; - } -} - /* Allow multiple parallel connections from a single client. */ static int file_can_multi_conn (void *handle) @@ -275,374 +154,6 @@ file_can_multi_conn (void *handle) return 1; } -static int -file_can_trim (void *handle) -{ - /* Trim is advisory, but we prefer to advertise it only when we can - * actually (attempt to) punch holes. Since not all filesystems - * support all fallocate modes, it would be nice if we had a way - * from fpathconf() to definitively learn what will work on a given -...

There's a lot of code in nbdkit-file-plugin which it would be nice to reuse elsewhere. One possible approach (as outlined here) is simply to move the file callbacks (like file.pread, file.pwrite, file.zero etc) to a new mini-library. They can then be consumed by other plugins fairly easily by doing: static void * foo_open (int readonly) { struct fileops *fops; int fd, flags; /*

Needs some work still, see in particular the commit message for patch 3. Rich.

Since v1: - rework .can_cache to be tri-state, with default of no advertisement (ripple effect through other patches) - add a lot more patches in order to round out filter support And in the meantime, Rich pushed NBD_CMD_CACHE support into libnbd, so in theory we now have a way to test cache commands through the entire stack. Eric Blake (24): server: Internal hooks for implementing