From d6e04ae64c6b06ef76a5d4fb49106b393b7fa50a Mon Sep 17 00:00:00 2001 From: Steve French Date: Mon, 13 Jun 2005 13:24:43 -0500 Subject: [PATCH] [CIFS] CIFS writepage improvements - eliminate double copy Signed-off-by: Steve French (sfrench@us.ibm.com) --- fs/cifs/cifsproto.h | 6 +- fs/cifs/cifssmb.c | 57 ++++++----- fs/cifs/file.c | 21 ++++- fs/cifs/transport.c | 223 ++++++++++++++++++++++++++++++++++---------- 4 files changed, 231 insertions(+), 76 deletions(-) diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index ea239dea571e..b43ac9230eab 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -47,6 +47,10 @@ extern int SendReceive(const unsigned int /* xid */ , struct cifsSesInfo *, struct smb_hdr * /* input */ , struct smb_hdr * /* out */ , int * /* bytes returned */ , const int long_op); +extern int SendReceive2(const unsigned int /* xid */ , struct cifsSesInfo *, + struct smb_hdr * /* input */ , int hdr_len, + const char * /* SMB data to send */ , int data_len, + int * /* bytes returned */ , const int long_op); extern int checkSMBhdr(struct smb_hdr *smb, __u16 mid); extern int checkSMB(struct smb_hdr *smb, __u16 mid, int length); extern int is_valid_oplock_break(struct smb_hdr *smb); @@ -222,7 +226,7 @@ extern int CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, extern int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, const __u64 offset, unsigned int *nbytes, - const char __user *buf,const int long_op); + const char *buf,const int long_op); extern int CIFSGetSrvInodeNumber(const int xid, struct cifsTconInfo *tcon, const unsigned char *searchName, __u64 * inode_number, const struct nls_table *nls_codepage, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 3c628bf667a5..b4f7b9859e3b 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -951,56 +951,69 @@ CIFSSMBWrite(const int xid, struct cifsTconInfo *tcon, } #ifdef CONFIG_CIFS_EXPERIMENTAL -int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, +int +CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, const int netfid, const unsigned int count, - const __u64 offset, unsigned int *nbytes, const char __user *buf, + const __u64 offset, unsigned int *nbytes, const char *buf, const int long_op) { int rc = -EACCES; WRITE_REQ *pSMB = NULL; - WRITE_RSP *pSMBr = NULL; - /*int bytes_returned;*/ - unsigned bytes_sent; + int bytes_returned; + int smb_hdr_len; + __u32 bytes_sent; __u16 byte_count; + cERROR(1,("write2 at %lld %d bytes",offset,count)); /* BB removeme BB */ rc = small_smb_init(SMB_COM_WRITE_ANDX, 14, tcon, (void **) &pSMB); - if (rc) return rc; - - pSMBr = (WRITE_RSP *)pSMB; /* BB removeme BB */ - /* tcon and ses pointer are checked in smb_init */ if (tcon->ses->server == NULL) return -ECONNABORTED; - pSMB->AndXCommand = 0xFF; /* none */ + pSMB->AndXCommand = 0xFF; /* none */ pSMB->Fid = netfid; pSMB->OffsetLow = cpu_to_le32(offset & 0xFFFFFFFF); pSMB->OffsetHigh = cpu_to_le32(offset >> 32); pSMB->Reserved = 0xFFFFFFFF; pSMB->WriteMode = 0; pSMB->Remaining = 0; - bytes_sent = (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE) & ~0xFF; + + /* Can increase buffer size if buffer is big enough in some cases - ie + can send more if LARGE_WRITE_X capability returned by the server and if + our buffer is big enough or if we convert to iovecs on socket writes + and eliminate the copy to the CIFS buffer */ + if(tcon->ses->capabilities & CAP_LARGE_WRITE_X) { + bytes_sent = min_t(const unsigned int, CIFSMaxBufSize, count); + } else { + bytes_sent = (tcon->ses->server->maxBuf - MAX_CIFS_HDR_SIZE) + & ~0xFF; + } + if (bytes_sent > count) bytes_sent = count; - pSMB->DataLengthHigh = 0; pSMB->DataOffset = cpu_to_le16(offsetof(struct smb_com_write_req,Data) - 4); - byte_count = bytes_sent + 1 /* pad */ ; - pSMB->DataLengthLow = cpu_to_le16(bytes_sent); - pSMB->DataLengthHigh = 0; - pSMB->hdr.smb_buf_length += byte_count; + byte_count = bytes_sent + 1 /* pad */ ; /* BB fix this for sends > 64K */ + pSMB->DataLengthLow = cpu_to_le16(bytes_sent & 0xFFFF); + pSMB->DataLengthHigh = cpu_to_le16(bytes_sent >> 16); + smb_hdr_len = pSMB->hdr.smb_buf_length + 1; /* hdr + 1 byte pad */ + pSMB->hdr.smb_buf_length += bytes_sent+1; pSMB->ByteCount = cpu_to_le16(byte_count); -/* rc = SendReceive2(xid, tcon->ses, (struct smb_hdr *) pSMB, - (struct smb_hdr *) pSMBr, buf, buflen, &bytes_returned, long_op); */ /* BB fixme BB */ + rc = SendReceive2(xid, tcon->ses, (struct smb_hdr *) pSMB, smb_hdr_len, + buf, bytes_sent, &bytes_returned, long_op); if (rc) { - cFYI(1, ("Send error in write2 (large write) = %d", rc)); + cFYI(1, ("Send error in write = %d", rc)); *nbytes = 0; - } else - *nbytes = le16_to_cpu(pSMBr->Count); + } else { + WRITE_RSP * pSMBr = (WRITE_RSP *)pSMB; + *nbytes = le16_to_cpu(pSMBr->CountHigh); + *nbytes = (*nbytes) << 16; + *nbytes += le16_to_cpu(pSMBr->Count); + } cifs_small_buf_release(pSMB); @@ -1009,6 +1022,8 @@ int CIFSSMBWrite2(const int xid, struct cifsTconInfo *tcon, return rc; } + + #endif /* CIFS_EXPERIMENTAL */ int diff --git a/fs/cifs/file.c b/fs/cifs/file.c index dde2d251fc3d..ca74c1151be9 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -791,9 +791,8 @@ static ssize_t cifs_write(struct file *file, const char *write_data, pTcon = cifs_sb->tcon; - /* cFYI(1, - (" write %d bytes to offset %lld of %s", write_size, - *poffset, file->f_dentry->d_name.name)); */ + cFYI(1,(" write %d bytes to offset %lld of %s", write_size, + *poffset, file->f_dentry->d_name.name)); /* BB removeme BB */ if (file->private_data == NULL) return -EBADF; @@ -846,7 +845,21 @@ static ssize_t cifs_write(struct file *file, const char *write_data, if (rc != 0) break; } - +#ifdef CIFS_EXPERIMENTAL + /* BB FIXME We can not sign across two buffers yet */ + cERROR(1,("checking signing")); /* BB removeme BB */ + if(pTcon->ses->server->secMode & + (SECMODE_SIGN_REQUIRED | SECMODE_SIGN_ENABLED) == 0) + rc = CIFSSMBWrite2(xid, pTcon, + open_file->netfid, + min_t(const int, cifs_sb->wsize, + write_size - total_written), + *poffset, &bytes_written, + write_data + total_written, + long_op); + } else + /* BB FIXME fixup indentation of line below */ +#endif rc = CIFSSMBWrite(xid, pTcon, open_file->netfid, min_t(const int, cifs_sb->wsize, diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 0046c219833d..04f4af07fdd4 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -49,7 +49,8 @@ AllocMidQEntry(struct smb_hdr *smb_buffer, struct cifsSesInfo *ses) return NULL; } - temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp,SLAB_KERNEL | SLAB_NOFS); + temp = (struct mid_q_entry *) mempool_alloc(cifs_mid_poolp, + SLAB_KERNEL | SLAB_NOFS); if (temp == NULL) return temp; else { @@ -179,27 +180,24 @@ smb_send(struct socket *ssocket, struct smb_hdr *smb_buffer, return rc; } -#ifdef CIFS_EXPERIMENTAL -/* BB finish off this function, adding support for writing set of pages as iovec */ -/* and also adding support for operations that need to parse the response smb */ - -int -smb_sendv(struct socket *ssocket, struct smb_hdr *smb_buffer, - unsigned int smb_buf_length, struct kvec * write_vector - /* page list */, struct sockaddr *sin) +#ifdef CONFIG_CIFS_EXPERIMENTAL +static int +smb_send2(struct socket *ssocket, struct smb_hdr *smb_buffer, + unsigned int smb_hdr_length, const char * data, unsigned int datalen, + struct sockaddr *sin) { int rc = 0; int i = 0; struct msghdr smb_msg; - number_of_pages += 1; /* account for SMB header */ - struct kvec * piov = kmalloc(number_of_pages * sizeof(struct kvec)); - unsigned len = smb_buf_length + 4; - + struct kvec iov[2]; + unsigned len = smb_hdr_length + 4; + if(ssocket == NULL) return -ENOTSOCK; /* BB eventually add reconnect code here */ - iov.iov_base = smb_buffer; - iov.iov_len = len; - + iov[0].iov_base = smb_buffer; + iov[0].iov_len = len; + iov[1].iov_base = data; + iov[2].iov_len = datalen; smb_msg.msg_name = sin; smb_msg.msg_namelen = sizeof (struct sockaddr); smb_msg.msg_control = NULL; @@ -212,12 +210,11 @@ smb_sendv(struct socket *ssocket, struct smb_hdr *smb_buffer, Flags2 is converted in SendReceive */ smb_buffer->smb_buf_length = cpu_to_be32(smb_buffer->smb_buf_length); - cFYI(1, ("Sending smb of length %d ", smb_buf_length)); + cFYI(1, ("Sending smb of length %d ", len + datalen)); dump_smb(smb_buffer, len); - while (len > 0) { - rc = kernel_sendmsg(ssocket, &smb_msg, &iov, number_of_pages, - len); + while (len + datalen > 0) { + rc = kernel_sendmsg(ssocket, &smb_msg, iov, 2, len); if ((rc == -ENOSPC) || (rc == -EAGAIN)) { i++; if(i > 60) { @@ -232,9 +229,22 @@ smb_sendv(struct socket *ssocket, struct smb_hdr *smb_buffer, } if (rc < 0) break; - iov.iov_base += rc; - iov.iov_len -= rc; - len -= rc; + if(iov[0].iov_len > 0) { + if(rc >= len) { + iov[0].iov_len = 0; + rc -= len; + } else { /* some of hdr was not sent */ + len -= rc; + iov[0].iov_len -= rc; + iov[0].iov_base += rc; + continue; + } + } + if((iov[0].iov_len == 0) && (rc > 0)){ + iov[1].iov_base += rc; + iov[1].iov_len -= rc; + datalen -= rc; + } } if (rc < 0) { @@ -246,14 +256,15 @@ smb_sendv(struct socket *ssocket, struct smb_hdr *smb_buffer, return rc; } - int -CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, - struct smb_hdr *in_buf, struct kvec * write_vector /* page list */, int *pbytes_returned, const int long_op) +SendReceive2(const unsigned int xid, struct cifsSesInfo *ses, + struct smb_hdr *in_buf, int hdrlen, const char * data, + int datalen, int *pbytes_returned, const int long_op) { int rc = 0; - unsigned long timeout = 15 * HZ; - struct mid_q_entry *midQ = NULL; + unsigned int receive_len; + unsigned long timeout; + struct mid_q_entry *midQ; if (ses == NULL) { cERROR(1,("Null smb session")); @@ -263,14 +274,8 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, cERROR(1,("Null tcp session")); return -EIO; } - if(pbytes_returned == NULL) - return -EIO; - else - *pbytes_returned = 0; - - - if(ses->server->tcpStatus == CIFS_EXITING) + if(ses->server->tcpStatus == CifsExiting) return -ENOENT; /* Ensure that we do not send more than 50 overlapping requests @@ -282,7 +287,8 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, } else { spin_lock(&GlobalMid_Lock); while(1) { - if(atomic_read(&ses->server->inFlight) >= cifs_max_pending){ + if(atomic_read(&ses->server->inFlight) >= + cifs_max_pending){ spin_unlock(&GlobalMid_Lock); wait_event(ses->server->request_q, atomic_read(&ses->server->inFlight) @@ -314,17 +320,17 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, if (ses->server->tcpStatus == CifsExiting) { rc = -ENOENT; - goto cifs_out_label; + goto out_unlock2; } else if (ses->server->tcpStatus == CifsNeedReconnect) { cFYI(1,("tcp session dead - return to caller to retry")); rc = -EAGAIN; - goto cifs_out_label; + goto out_unlock2; } else if (ses->status != CifsGood) { /* check if SMB session is bad because we are setting it up */ if((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && (in_buf->Command != SMB_COM_NEGOTIATE)) { rc = -EAGAIN; - goto cifs_out_label; + goto out_unlock2; } /* else ok - we are setting up session */ } midQ = AllocMidQEntry(in_buf, ses); @@ -352,13 +358,12 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, return -EIO; } - /* BB can we sign efficiently in this path? */ - rc = cifs_sign_smb(in_buf, ses->server, &midQ->sequence_number); +/* BB FIXME */ +/* rc = cifs_sign_smb2(in_buf, data, ses->server, &midQ->sequence_number); */ midQ->midState = MID_REQUEST_SUBMITTED; -/* rc = smb_sendv(ses->server->ssocket, in_buf, in_buf->smb_buf_length, - piovec, - (struct sockaddr *) &(ses->server->addr.sockAddr));*/ + rc = smb_send2(ses->server->ssocket, in_buf, hdrlen, data, datalen, + (struct sockaddr *) &(ses->server->addr.sockAddr)); if(rc < 0) { DeleteMidQEntry(midQ); up(&ses->server->tcpSem); @@ -370,19 +375,137 @@ CIFSSendRcv(const unsigned int xid, struct cifsSesInfo *ses, return rc; } else up(&ses->server->tcpSem); -cifs_out_label: - if(midQ) - DeleteMidQEntry(midQ); - + if (long_op == -1) + goto cifs_no_response_exit2; + else if (long_op == 2) /* writes past end of file can take loong time */ + timeout = 300 * HZ; + else if (long_op == 1) + timeout = 45 * HZ; /* should be greater than + servers oplock break timeout (about 43 seconds) */ + else if (long_op > 2) { + timeout = MAX_SCHEDULE_TIMEOUT; + } else + timeout = 15 * HZ; + /* wait for 15 seconds or until woken up due to response arriving or + due to last connection to this server being unmounted */ + if (signal_pending(current)) { + /* if signal pending do not hold up user for full smb timeout + but we still give response a change to complete */ + timeout = 2 * HZ; + } + + /* No user interrupts in wait - wreaks havoc with performance */ + if(timeout != MAX_SCHEDULE_TIMEOUT) { + timeout += jiffies; + wait_event(ses->server->response_q, + (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + time_after(jiffies, timeout) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + } else { + wait_event(ses->server->response_q, + (!(midQ->midState & MID_REQUEST_SUBMITTED)) || + ((ses->server->tcpStatus != CifsGood) && + (ses->server->tcpStatus != CifsNew))); + } + + spin_lock(&GlobalMid_Lock); + if (midQ->resp_buf) { + spin_unlock(&GlobalMid_Lock); + receive_len = be32_to_cpu(*(__be32 *)midQ->resp_buf); + } else { + cERROR(1,("No response buffer")); + if(midQ->midState == MID_REQUEST_SUBMITTED) { + if(ses->server->tcpStatus == CifsExiting) + rc = -EHOSTDOWN; + else { + ses->server->tcpStatus = CifsNeedReconnect; + midQ->midState = MID_RETRY_NEEDED; + } + } + + if (rc != -EHOSTDOWN) { + if(midQ->midState == MID_RETRY_NEEDED) { + rc = -EAGAIN; + cFYI(1,("marking request for retry")); + } else { + rc = -EIO; + } + } + spin_unlock(&GlobalMid_Lock); + DeleteMidQEntry(midQ); + /* If not lock req, update # of requests on wire to server */ + if(long_op < 3) { + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); + } + return rc; + } + + if (receive_len > CIFSMaxBufSize + MAX_CIFS_HDR_SIZE) { + cERROR(1, ("Frame too large received. Length: %d Xid: %d", + receive_len, xid)); + rc = -EIO; + } else { /* rcvd frame is ok */ + + if (midQ->resp_buf && + (midQ->midState == MID_RESPONSE_RECEIVED)) { + in_buf->smb_buf_length = receive_len; + /* BB verify that length would not overrun small buf */ + memcpy((char *)in_buf + 4, + (char *)midQ->resp_buf + 4, + receive_len); + + dump_smb(in_buf, 80); + /* convert the length into a more usable form */ + if((receive_len > 24) && + (ses->server->secMode & (SECMODE_SIGN_REQUIRED | + SECMODE_SIGN_ENABLED))) { + rc = cifs_verify_signature(in_buf, + ses->server->mac_signing_key, + midQ->sequence_number+1); + if(rc) { + cERROR(1,("Unexpected SMB signature")); + /* BB FIXME add code to kill session */ + } + } + + *pbytes_returned = in_buf->smb_buf_length; + + /* BB special case reconnect tid and uid here? */ + rc = map_smb_to_linux_error(in_buf); + + /* convert ByteCount if necessary */ + if (receive_len >= + sizeof (struct smb_hdr) - + 4 /* do not count RFC1001 header */ + + (2 * in_buf->WordCount) + 2 /* bcc */ ) + BCC(in_buf) = le16_to_cpu(BCC(in_buf)); + } else { + rc = -EIO; + cFYI(1,("Bad MID state? ")); + } + } +cifs_no_response_exit2: + DeleteMidQEntry(midQ); + if(long_op < 3) { - atomic_dec(&ses->server->inFlight); + atomic_dec(&ses->server->inFlight); wake_up(&ses->server->request_q); } return rc; -} +out_unlock2: + up(&ses->server->tcpSem); + /* If not lock req, update # of requests on wire to server */ + if(long_op < 3) { + atomic_dec(&ses->server->inFlight); + wake_up(&ses->server->request_q); + } + return rc; +} #endif /* CIFS_EXPERIMENTAL */ int -- 2.39.2