qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Subject: [PULL 18/28] file-posix: try BLKSECTGET on block devices too, do not round to power of 2
Date: Fri, 25 Jun 2021 16:18:12 +0200	[thread overview]
Message-ID: <20210625141822.1368639-19-pbonzini@redhat.com> (raw)
In-Reply-To: <20210625141822.1368639-1-pbonzini@redhat.com>

bs->sg is only true for character devices, but block devices can also
be used with scsi-block and scsi-generic.  Unfortunately BLKSECTGET
returns bytes in an int for /dev/sgN devices, and sectors in a short
for block devices, so account for that in the code.

The maximum transfer also need not be a power of 2 (for example I have
seen disks with 1280 KiB maximum transfer) so there's no need to pass
the result through pow2floor.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 block/file-posix.c | 67 ++++++++++++++++++++++++++--------------------
 1 file changed, 38 insertions(+), 29 deletions(-)

diff --git a/block/file-posix.c b/block/file-posix.c
index 88e58d2863..ea102483b0 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
@@ -1147,22 +1147,27 @@ static void raw_reopen_abort(BDRVReopenState *state)
     s->reopen_state = NULL;
 }
 
-static int sg_get_max_transfer_length(int fd)
+static int hdev_get_max_hw_transfer(int fd, struct stat *st)
 {
 #ifdef BLKSECTGET
-    int max_bytes = 0;
-
-    if (ioctl(fd, BLKSECTGET, &max_bytes) == 0) {
-        return max_bytes;
+    if (S_ISBLK(st->st_mode)) {
+        unsigned short max_sectors = 0;
+        if (ioctl(fd, BLKSECTGET, &max_sectors) == 0) {
+            return max_sectors * 512;
+        }
     } else {
-        return -errno;
+        int max_bytes = 0;
+        if (ioctl(fd, BLKSECTGET, &max_bytes) == 0) {
+            return max_bytes;
+        }
     }
+    return -errno;
 #else
     return -ENOSYS;
 #endif
 }
 
-static int sg_get_max_segments(int fd)
+static int hdev_get_max_segments(int fd, struct stat *st)
 {
 #ifdef CONFIG_LINUX
     char buf[32];
@@ -1171,26 +1176,20 @@ static int sg_get_max_segments(int fd)
     int ret;
     int sysfd = -1;
     long max_segments;
-    struct stat st;
 
-    if (fstat(fd, &st)) {
-        ret = -errno;
-        goto out;
-    }
-
-    if (S_ISCHR(st.st_mode)) {
+    if (S_ISCHR(st->st_mode)) {
         if (ioctl(fd, SG_GET_SG_TABLESIZE, &ret) == 0) {
             return ret;
         }
         return -ENOTSUP;
     }
 
-    if (!S_ISBLK(st.st_mode)) {
+    if (!S_ISBLK(st->st_mode)) {
         return -ENOTSUP;
     }
 
     sysfspath = g_strdup_printf("/sys/dev/block/%u:%u/queue/max_segments",
-                                major(st.st_rdev), minor(st.st_rdev));
+                                major(st->st_rdev), minor(st->st_rdev));
     sysfd = open(sysfspath, O_RDONLY);
     if (sysfd == -1) {
         ret = -errno;
@@ -1227,23 +1226,33 @@ out:
 static void raw_refresh_limits(BlockDriverState *bs, Error **errp)
 {
     BDRVRawState *s = bs->opaque;
-
-    if (bs->sg) {
-        int ret = sg_get_max_transfer_length(s->fd);
-
-        if (ret > 0 && ret <= BDRV_REQUEST_MAX_BYTES) {
-            bs->bl.max_hw_transfer = pow2floor(ret);
-        }
-
-        ret = sg_get_max_segments(s->fd);
-        if (ret > 0) {
-            bs->bl.max_iov = ret;
-        }
-    }
+    struct stat st;
 
     raw_probe_alignment(bs, s->fd, errp);
     bs->bl.min_mem_alignment = s->buf_align;
     bs->bl.opt_mem_alignment = MAX(s->buf_align, qemu_real_host_page_size);
+
+    /*
+     * Maximum transfers are best effort, so it is okay to ignore any
+     * errors.  That said, based on the man page errors in fstat would be
+     * very much unexpected; the only possible case seems to be ENOMEM.
+     */
+    if (fstat(s->fd, &st)) {
+        return;
+    }
+
+    if (bs->sg || S_ISBLK(st.st_mode)) {
+        int ret = hdev_get_max_hw_transfer(s->fd, &st);
+
+        if (ret > 0 && ret <= BDRV_REQUEST_MAX_BYTES) {
+            bs->bl.max_hw_transfer = ret;
+        }
+
+        ret = hdev_get_max_segments(s->fd, &st);
+        if (ret > 0) {
+            bs->bl.max_iov = ret;
+        }
+    }
 }
 
 static int check_for_dasd(int fd)
-- 
2.31.1




  parent reply	other threads:[~2021-06-25 14:35 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-25 14:17 [PULL v2 00/28] Misc (including block file-posix) for 2021-06-23 Paolo Bonzini
2021-06-25 14:17 ` [PULL 01/28] target/i386: kvm: add support for TSC scaling Paolo Bonzini
2021-06-25 14:17 ` [PULL 02/28] meson: drop unused CONFIG_GCRYPT_HMAC Paolo Bonzini
2021-06-25 14:17 ` [PULL 03/28] configure: drop unused variables for xts Paolo Bonzini
2021-06-25 14:17 ` [PULL 04/28] meson: remove preadv from summary Paolo Bonzini
2021-06-25 14:17 ` [PULL 05/28] tests: remove QCRYPTO_HAVE_TLS_TEST_SUPPORT Paolo Bonzini
2021-06-25 14:18 ` [PULL 06/28] configure, meson: convert crypto detection to meson Paolo Bonzini
2021-06-25 14:18 ` [PULL 07/28] configure, meson: convert libtasn1 " Paolo Bonzini
2021-06-25 14:18 ` [PULL 08/28] configure, meson: convert pam " Paolo Bonzini
2021-06-25 14:18 ` [PULL 09/28] configure, meson: convert libusb " Paolo Bonzini
2021-06-25 14:18 ` [PULL 10/28] configure, meson: convert libcacard " Paolo Bonzini
2021-06-25 14:18 ` [PULL 11/28] configure, meson: convert libusbredir " Paolo Bonzini
2021-06-25 14:18 ` [PULL 12/28] KVM: Fix dirty ring mmap incorrect size due to renaming accident Paolo Bonzini
2021-06-25 14:18 ` [PULL 13/28] file-posix: fix max_iov for /dev/sg devices Paolo Bonzini
2021-06-25 14:18 ` [PULL 14/28] scsi-generic: pass max_segments via max_iov field in BlockLimits Paolo Bonzini
2021-06-25 14:18 ` [PULL 15/28] osdep: provide ROUND_DOWN macro Paolo Bonzini
2021-06-29  4:12   ` Philippe Mathieu-Daudé
2021-06-25 14:18 ` [PULL 16/28] block-backend: align max_transfer to request alignment Paolo Bonzini
2021-06-25 14:18 ` [PULL 17/28] block: add max_hw_transfer to BlockLimits Paolo Bonzini
2021-06-25 14:18 ` Paolo Bonzini [this message]
2021-09-06 14:24   ` [PULL 18/28] file-posix: try BLKSECTGET on block devices too, do not round to power of 2 Halil Pasic
2021-09-22 19:51     ` Halil Pasic
2021-09-23  9:18       ` Recent qemu patch results in aio failures with host DASD disks resulting in guest I/O errors Christian Borntraeger
2021-09-23 10:57       ` [PULL 18/28] file-posix: try BLKSECTGET on block devices too, do not round to power of 2 Paolo Bonzini
2021-09-23 12:13         ` Halil Pasic
2021-09-23 13:02           ` Paolo Bonzini
2021-06-25 14:18 ` [PULL 19/28] block: feature detection for host block support Paolo Bonzini
2021-06-25 14:18 ` [PULL 20/28] block: check for sys/disk.h Paolo Bonzini
2021-06-25 14:18 ` [PULL 21/28] block: try BSD disk size ioctls one after another Paolo Bonzini
2021-06-25 14:18 ` [PULL 22/28] block: detect DKIOCGETBLOCKCOUNT/SIZE before use Paolo Bonzini
2021-06-25 14:18 ` [PULL 23/28] file-posix: handle EINTR during ioctl Paolo Bonzini
2021-06-25 14:18 ` [PULL 24/28] machine: move dies from X86MachineState to CpuTopology Paolo Bonzini
2021-06-25 14:18 ` [PULL 25/28] machine: move common smp_parse code to caller Paolo Bonzini
2021-06-25 14:18 ` [PULL 26/28] machine: add error propagation to mc->smp_parse Paolo Bonzini
2021-06-25 14:18 ` [PULL 27/28] machine: pass QAPI struct " Paolo Bonzini
2021-06-25 14:18 ` [PULL 28/28] machine: reject -smp dies!=1 for non-PC machines Paolo Bonzini
2021-06-29  8:37 ` [PULL v2 00/28] Misc (including block file-posix) for 2021-06-23 Peter Maydell

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210625141822.1368639-19-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).