From: Greg KH <gregkh@suse.de>
To: linux-kernel@vger.kernel.org, stable@kernel.org
Cc: stable-review@kernel.org, torvalds@linux-foundation.org,
akpm@linux-foundation.org, alan@lxorguk.ukuu.org.uk,
Jeff Layton <jlayton@redhat.com>,
Steve French <sfrench@us.ibm.com>
Subject: [62/70] cifs: clean up wsize negotiation and allow for larger wsize
Date: Mon, 01 Aug 2011 16:19:17 -0700 [thread overview]
Message-ID: <20110801232056.812072156@clark.kroah.org> (raw)
In-Reply-To: <20110801232124.GA15313@kroah.com>
2.6.39-stable review patch. If anyone has any objections, please let us know.
------------------
From: Jeff Layton <jlayton@redhat.com>
commit f7910cbd9fa319ee4501074f1f3b5ce23c4b1518 upstream.
Now that we can handle larger wsizes in writepages, fix up the
negotiation of the wsize to allow for that. find_get_pages only seems to
give out a max of 256 pages at a time, so that gives us a reasonable
default of 1M for the wsize.
If the server however does not support large writes via POSIX
extensions, then we cap the wsize to (128k - PAGE_CACHE_SIZE). That
gives us a size that goes up to the max frame size specified in RFC1001.
Finally, if CAP_LARGE_WRITE_AND_X isn't set, then further cap it to the
largest size allowed by the protocol (USHRT_MAX).
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Reviewed-and-Tested-by: Pavel Shilovsky <piastry@etersoft.ru>
Signed-off-by: Steve French <sfrench@us.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
fs/cifs/connect.c | 69 ++++++++++++++++++++++++++++++++++++++----------------
1 file changed, 49 insertions(+), 20 deletions(-)
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -2571,23 +2571,6 @@ static void setup_cifs_sb(struct smb_vol
else /* default */
cifs_sb->rsize = CIFSMaxBufSize;
- if (pvolume_info->wsize > PAGEVEC_SIZE * PAGE_CACHE_SIZE) {
- cERROR(1, "wsize %d too large, using 4096 instead",
- pvolume_info->wsize);
- cifs_sb->wsize = 4096;
- } else if (pvolume_info->wsize)
- cifs_sb->wsize = pvolume_info->wsize;
- else
- cifs_sb->wsize = min_t(const int,
- PAGEVEC_SIZE * PAGE_CACHE_SIZE,
- 127*1024);
- /* old default of CIFSMaxBufSize was too small now
- that SMB Write2 can send multiple pages in kvec.
- RFC1001 does not describe what happens when frame
- bigger than 128K is sent so use that as max in
- conjunction with 52K kvec constraint on arch with 4K
- page size */
-
if (cifs_sb->rsize < 2048) {
cifs_sb->rsize = 2048;
/* Windows ME may prefer this */
@@ -2665,6 +2648,53 @@ static void setup_cifs_sb(struct smb_vol
"mount option supported");
}
+/*
+ * When the server supports very large writes via POSIX extensions, we can
+ * allow up to 2^24 - PAGE_CACHE_SIZE.
+ *
+ * Note that this might make for "interesting" allocation problems during
+ * writeback however (as we have to allocate an array of pointers for the
+ * pages). A 16M write means ~32kb page array with PAGE_CACHE_SIZE == 4096.
+ */
+#define CIFS_MAX_WSIZE ((1<<24) - PAGE_CACHE_SIZE)
+
+/*
+ * When the server doesn't allow large posix writes, default to a wsize of
+ * 128k - PAGE_CACHE_SIZE -- one page less than the largest frame size
+ * described in RFC1001. This allows space for the header without going over
+ * that by default.
+ */
+#define CIFS_MAX_RFC1001_WSIZE (128 * 1024 - PAGE_CACHE_SIZE)
+
+/*
+ * The default wsize is 1M. find_get_pages seems to return a maximum of 256
+ * pages in a single call. With PAGE_CACHE_SIZE == 4k, this means we can fill
+ * a single wsize request with a single call.
+ */
+#define CIFS_DEFAULT_WSIZE (1024 * 1024)
+
+static unsigned int
+cifs_negotiate_wsize(struct cifsTconInfo *tcon, struct smb_vol *pvolume_info)
+{
+ __u64 unix_cap = le64_to_cpu(tcon->fsUnixInfo.Capability);
+ struct TCP_Server_Info *server = tcon->ses->server;
+ unsigned int wsize = pvolume_info->wsize ? pvolume_info->wsize :
+ CIFS_DEFAULT_WSIZE;
+
+ /* can server support 24-bit write sizes? (via UNIX extensions) */
+ if (!tcon->unix_ext || !(unix_cap & CIFS_UNIX_LARGE_WRITE_CAP))
+ wsize = min_t(unsigned int, wsize, CIFS_MAX_RFC1001_WSIZE);
+
+ /* no CAP_LARGE_WRITE_X? Limit it to 16 bits */
+ if (!(server->capabilities & CAP_LARGE_WRITE_X))
+ wsize = min_t(unsigned int, wsize, USHRT_MAX);
+
+ /* hard limit of CIFS_MAX_WSIZE */
+ wsize = min_t(unsigned int, wsize, CIFS_MAX_WSIZE);
+
+ return wsize;
+}
+
static int
is_path_accessible(int xid, struct cifsTconInfo *tcon,
struct cifs_sb_info *cifs_sb, const char *full_path)
@@ -2866,13 +2896,12 @@ try_mount_again:
cifs_sb->rsize = 1024 * 127;
cFYI(DBG2, "no very large read support, rsize now 127K");
}
- if (!(tcon->ses->capabilities & CAP_LARGE_WRITE_X))
- cifs_sb->wsize = min(cifs_sb->wsize,
- (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE));
if (!(tcon->ses->capabilities & CAP_LARGE_READ_X))
cifs_sb->rsize = min(cifs_sb->rsize,
(tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE));
+ cifs_sb->wsize = cifs_negotiate_wsize(tcon, volume_info);
+
remote_path_check:
/* check if a whole path (including prepath) is not remote */
if (!rc && tcon) {
next prev parent reply other threads:[~2011-08-01 23:28 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-08-01 23:21 [00/70] 2.6.39.4-stable review Greg KH
2011-08-01 23:18 ` [01/70] USB: serial: add IDs for WinChipHead USB->RS232 adapter Greg KH
2011-08-01 23:18 ` [02/70] drm/radeon/kms: free ib pool on module unloading Greg KH
2011-08-01 23:18 ` [03/70] drm/radeon/kms: fix typo in evergreen disp int status register Greg KH
2011-08-01 23:18 ` [04/70] drm/radeon/kms: fix typo in IH_CNTL swap bitfield Greg KH
2011-08-01 23:18 ` [05/70] mm: vmscan: correct check for kswapd sleeping in sleeping_prematurely Greg KH
2011-08-01 23:18 ` [06/70] mm/nommu.c: fix remap_pfn_range() Greg KH
2011-08-01 23:18 ` [07/70] mm: vmscan: evaluate the watermarks against the correct classzone Greg KH
2011-08-01 23:18 ` [08/70] hwmon: (lm95241) Fix chip detection code Greg KH
2011-08-01 23:18 ` [09/70] hwmon: (lm95241) Fix negative temperature results Greg KH
2011-08-01 23:18 ` [10/70] hwmon: (pmbus) Auto-detect temp2 and temp3 registers/attributes Greg KH
2011-08-01 23:18 ` [11/70] hwmon: (pmbus) Drop check for PMBus revision register in probe function Greg KH
2011-08-01 23:18 ` [12/70] hwmon: (pmbus) Improve fan detection Greg KH
2011-08-01 23:18 ` [13/70] hwmon: (pmbus) Use long variables for register to data conversions Greg KH
2011-08-01 23:18 ` [14/70] ASoC: Fix Blackfin I2S _pointer() implementation return in bounds values Greg KH
2011-08-01 23:18 ` [15/70] ASoC: Ensure we delay long enough for WM8994 FLL to lock when starting Greg KH
2011-08-01 23:18 ` [16/70] ASoC: ak4642: fixup snd_soc_update_bits mask for PW_MGMT2 Greg KH
2011-08-01 23:18 ` [17/70] ARM: 6989/1: perf: do not start the PMU when no events are present Greg KH
2011-08-01 23:18 ` [18/70] ARM: pxa168: correct nand pmu setting Greg KH
2011-08-01 23:18 ` [19/70] ARM: pxa910: " Greg KH
2011-08-01 23:18 ` [20/70] ARM: pxa: fix PGSR register address calculation Greg KH
2011-08-01 23:18 ` [21/70] firewire: ohci: do not bind to Pinnacle cards, avert panic Greg KH
2011-08-01 23:18 ` [22/70] Revert: "dell-laptop: Toggle the unsupported hardware killswitch" Greg KH
2011-08-01 23:18 ` [23/70] [media] v4l2-ioctl.c: prefill tuner type for g_frequency and g/s_tuner Greg KH
2011-08-01 23:18 ` [24/70] [media] tuner-core: fix s_std and s_tuner Greg KH
2011-08-01 23:18 ` [25/70] [media] tuner-core: fix tuner_resume: use t->mode instead of t->type Greg KH
2011-08-01 23:18 ` [26/70] hwmon: (pmbus) Improve auto-detection of temperature status register Greg KH
2011-08-01 23:18 ` [27/70] [media] pvrusb2: fix g/s_tuner support Greg KH
2011-08-01 23:18 ` [28/70] [media] bttv: fix s_tuner for radio Greg KH
2011-08-01 23:18 ` [29/70] fs/partitions/efi.c: corrupted GUID partition tables can cause kernel oops Greg KH
2011-08-01 23:18 ` [30/70] drm/radeon/kms: fix backend map typo on juniper Greg KH
2011-08-01 23:18 ` [31/70] NFSv4.1: update nfs4_fattr_bitmap_maxsz Greg KH
2011-08-01 23:18 ` [32/70] mac80211: fix TKIP replay vulnerability Greg KH
2011-08-01 23:18 ` [33/70] carl9170: add NEC WL300NU-AG usbid Greg KH
2011-08-01 23:18 ` [34/70] rtlwifi: rtl8192cu: Add new USB ID for Netgear WNA1000M Greg KH
2011-08-01 23:18 ` [35/70] SUNRPC: Fix a race between work-queue and rpc_killall_tasks Greg KH
2011-08-01 23:18 ` [36/70] SUNRPC: Fix use of static variable in rpcb_getport_async Greg KH
2011-08-01 23:18 ` [37/70] drm/radeon/kms: add new NI pci ids Greg KH
2011-08-01 23:18 ` [38/70] ath9k: Fix tx throughput drops for AR9003 chips with AES encryption Greg KH
2011-08-01 23:18 ` [39/70] ath5k: fix incorrect use of drvdata in sysfs code Greg KH
2011-08-01 23:18 ` [40/70] ath5k: fix incorrect use of drvdata in PCI suspend/resume code Greg KH
2011-08-01 23:18 ` [41/70] drm/radeon/kms: clean up multiple crtc handling for Greg KH
2011-08-01 23:18 ` [42/70] drm/radeon: fix oops in ttm reserve when pageflipping (v2) Greg KH
2011-08-01 23:18 ` [43/70] GFS2: make sure fallocate bytes is a multiple of blksize Greg KH
2011-08-01 23:18 ` [44/70] [media] Revert "V4L/DVB: cx23885: Enable Message Signaled Interrupts(MSI)" Greg KH
2011-08-01 23:19 ` [45/70] si4713-i2c: avoid potential buffer overflow on si4713 Greg KH
2011-08-01 23:19 ` [46/70] ASoC: Correct WM8994 MICBIAS supply widget hookup Greg KH
2011-08-01 23:19 ` [47/70] ASoC: Fix shift in WM8958 accessory detection default Greg KH
2011-08-01 23:19 ` [48/70] hwmon: (it87) Fix label group removal Greg KH
2011-08-01 23:19 ` [49/70] hwmon: (asus_atk0110) Fix memory leak Greg KH
2011-08-01 23:19 ` [50/70] hwmon: (max1111) Fix race condition causing NULL pointer exception Greg KH
2011-08-01 23:19 ` [51/70] mmc: Added quirks for Ricoh 1180:e823 lower base clock Greg KH
2011-08-01 23:19 ` [52/70] mmc: sdhci-esdhc-imx: SDHCI_CARD_PRESENT does not get Greg KH
2011-08-01 23:19 ` [53/70] x86: Make Dell Latitude E5420 use reboot=pci Greg KH
2011-08-01 23:19 ` [54/70] davinci: DM365 EVM: fix video input mux bits Greg KH
2011-08-01 23:19 ` [55/70] vmscan: fix a livelock in kswapd Greg KH
2011-08-01 23:19 ` [56/70] libata: fix unexpectedly frozen port after ata_eh_reset() Greg KH
2011-08-01 23:19 ` [57/70] ext4: fix oops in ext4_quota_off() Greg KH
2011-08-01 23:19 ` [58/70] mm: compaction: ensure that the compaction free scanner does not move to the next zone Greg KH
2011-08-01 23:19 ` [59/70] mm: vmscan: do not use page_count without a page pin Greg KH
2011-08-01 23:19 ` [60/70] [stable] [PATCH 3/3] mm: compaction: abort compaction if too many pages are isolated and caller is asynchronous V2 Greg KH
2011-08-01 23:19 ` [61/70] xfs [stable only]: restart busy extent search after node removal Greg KH
2011-08-01 23:19 ` Greg KH [this message]
2011-08-01 23:19 ` [63/70] mm: vmscan: do not apply pressure to slab if we are not applying pressure to zone Greg KH
2011-08-01 23:19 ` [64/70] mm: vmscan: only read new_classzone_idx from pgdat when reclaiming successfully Greg KH
2011-08-01 23:19 ` [65/70] Revert "DiBxxxx: get rid of DMA buffer on stack" Greg KH
2011-08-01 23:19 ` [66/70] vfs: fix race in rcu lookup of pruned dentry Greg KH
2011-08-01 23:19 ` [67/70] cifs: fix wsize negotiation to respect max buffer size and Greg KH
2011-08-01 23:19 ` [68/70] cifs: lower default and max wsize to what 2.6.39 can handle Greg KH
2011-08-01 23:19 ` [69/70] bridge: send proper message_age in config BPDU Greg KH
2011-08-01 23:19 ` [70/70] gro: Only reset frag0 when skb can be pulled Greg KH
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=20110801232056.812072156@clark.kroah.org \
--to=gregkh@suse.de \
--cc=akpm@linux-foundation.org \
--cc=alan@lxorguk.ukuu.org.uk \
--cc=jlayton@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=sfrench@us.ibm.com \
--cc=stable-review@kernel.org \
--cc=stable@kernel.org \
--cc=torvalds@linux-foundation.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
all inboxes | Powered by JetHome®