ip分片源碼解析（基於linux1.2.13）

fp = (struct ipfrag *) kmalloc(sizeof(struct ipfrag), GFP_ATOMIC);
if (fp == NULL)
{
printk("IP: frag_create: no memory left !
");
return(NULL);
}
memset(fp, 0, sizeof(struct ipfrag));

/* Fill in the structure. */
fp->offset = offset; // ip分配的首位元組在未分片數據中的偏移
fp->end = end; // 最後一個位元組的偏移 + 1，即下一個分片的首位元組偏移
fp->len = end - offset; // 分片長度
fp->skb = skb;
fp->ptr = ptr; // 指向分片的數據首地址

return(fp);
}
// 根據ip頭找到分片隊列的頭指針
static struct ipq *ip_find(struct iphdr *iph)
{
struct ipq *qp;
struct ipq *qplast;

cli();
qplast = NULL;
for(qp = ipqueue; qp != NULL; qplast = qp, qp = qp->next)
{ // 對比ip頭裡的幾個欄位
if (iph->id== qp->iph->id && iph->saddr == qp->iph->saddr &&
iph->daddr == qp->iph->daddr && iph->protocol == qp->iph->protocol)
{ // 找到後重置計時器，在這刪除，在ip_find外面新增一個計時
del_timer(&qp->timer); /* So it doesnt vanish on us. The timer will be reset anyway */
sti();
return(qp);
}
}
sti();
return(NULL);
}
// 釋放ip分片隊列
static void ip_free(struct ipq *qp)
{
struct ipfrag *fp;
struct ipfrag *xp;

/*
* Stop the timer for this entry.
*/
// 刪除定時器
del_timer(&qp->timer);

/* Remove this entry from the "incomplete datagrams" queue. */
cli();
/*
被刪除的節點前面沒有節點說明他是第一個節點，因為不是循環鏈表，
修改首指針ipqueue指向被刪除節點的下一個，如果下一個不為空，下一個節點的prev節點指向空，
因為這時候他為第一個節點。
*/
if (qp->prev == NULL)
{
ipqueue = qp->next;
if (ipqueue != NULL)
ipqueue->prev = NULL;
}
else
{
/*
被刪除節點不是第一個節點，但可能是最後一個，
被刪除節點的前一個節點的next指針指向被刪除節點的下一個節點，
如果如果被刪除節點的下一個節點不為空則他的prev指針執行被刪除節點
前面的節點
*/
qp->prev->next = qp->next;
if (qp->next != NULL)
qp->next->prev = qp->prev;
}

/* Release all fragment data. */

fp = qp->fragments;
// 刪除所有分片節點
while (fp != NULL)
{
xp = fp->next;
IS_SKB(fp->skb);
kfree_skb(fp->skb,FREE_READ);
kfree_s(fp, sizeof(struct ipfrag));
fp = xp;
}
// 刪除mac頭和ip頭，8位元組是icmp用的，存放傳輸層的前8個位元組
/* Release the MAC header. */
kfree_s(qp->mac, qp->maclen);

/* Release the IP header. */
kfree_s(qp->iph, qp->ihlen + 8);

/* Finally, release the queue descriptor itself. */
kfree_s(qp, sizeof(struct ipq));
sti();
}
// 分片重組超時處理函數
static void ip_expire(unsigned long arg)
{
struct ipq *qp;

qp = (struct ipq *)arg;

/*
* Send an ICMP "Fragment Reassembly Timeout" message.
*/

ip_statistics.IpReasmTimeout++;
ip_statistics.IpReasmFails++;
/* This if is always true... shrug */
// 發送icmp超時報文
if(qp->fragments!=NULL)
icmp_send(qp->fragments->skb,ICMP_TIME_EXCEEDED,
ICMP_EXC_FRAGTIME, 0, qp->dev);

/*
* Nuke the fragment queue.
*/
// 釋放分片隊列
ip_free(qp);
}
// 創建一個隊列用於重組分片
static struct ipq *ip_create(struct sk_buff *skb, struct iphdr *iph, struct device *dev)
{
struct ipq *qp;
int maclen;
int ihlen;
// 分片一個新的表示分片隊列的節點
qp = (struct ipq *) kmalloc(sizeof(struct ipq), GFP_ATOMIC);
if (qp == NULL)
{
printk("IP: create: no memory left !
");
return(NULL);
skb->dev = qp->dev;
}
memset(qp, 0, sizeof(struct ipq));

/*
* Allocate memory for the MAC header.
*
* FIXME: We have a maximum MAC address size limit and define
* elsewhere. We should use it here and avoid the 3 kmalloc() calls
*/
// mac頭長度等於ip頭減去mac頭首地址
maclen = ((unsigned long) iph) - ((unsigned long) skb->data);
qp->mac = (unsigned char *) kmalloc(maclen, GFP_ATOMIC);
if (qp->mac == NULL)
{
printk("IP: create: no memory left !
");
kfree_s(qp, sizeof(struct ipq));
return(NULL);
}

/*
* Allocate memory for the IP header (plus 8 octets for ICMP).
*/
// ip頭長度由ip頭欄位得出，多分配8個位元組給icmp
ihlen = (iph->ihl * sizeof(unsigned long));
qp->iph = (struct iphdr *) kmalloc(ihlen + 8, GFP_ATOMIC);
if (qp->iph == NULL)
{
printk("IP: create: no memory left !
");
kfree_s(qp->mac, maclen);
kfree_s(qp, sizeof(struct ipq));
return(NULL);
}

/* Fill in the structure. */
// 把mac頭內容複製到mac欄位
memcpy(qp->mac, skb->data, maclen);
// 把ip頭和傳輸層的8個位元組複製到iph欄位，8個欄位的內容用於發送icmp報文時
memcpy(qp->iph, iph, ihlen + 8);
// 未分片的ip報文的總長度，未知，收到所有分片後重新賦值
qp->len = 0;
// 當前分片的ip頭和mac頭長度
qp->ihlen = ihlen;
qp->maclen = maclen;
qp->fragments = NULL;
qp->dev = dev;

/* Start a timer for this entry. */
// 開始計時，一定時間內還沒收到所有分片則重組失敗，發送icmp報文
qp->timer.expires = IP_FRAG_TIME; /* about 30 seconds */
qp->timer.data = (unsigned long) qp; /* pointer to queue */
qp->timer.function = ip_expire; /* expire function */
add_timer(&qp->timer);

/* Add this entry to the queue. */
qp->prev = NULL;
cli();
// 頭插法插入分片重組的隊列
qp->next = ipqueue;
// 如果當前新增的節點不是第一個節點則把當前第一個節點的prev指針指向新增的節點
if (qp->next != NULL)
qp->next->prev = qp;
//更新ipqueue指向新增的節點，新增節點是首節點
ipqueue = qp;
sti();
return(qp);
}
// 判斷分片是否全部到達
static int ip_done(struct ipq *qp)
{
struct ipfrag *fp;
int offset;

/* Only possible if we received the final fragment. */
// 收到最後分片的時候會更新len欄位，如果沒有收到他就是初始化0，所以為0說明最後一個分片還沒到達，直接返回未完成
if (qp->len == 0)
return(0);
// 走到這裡說明全部分片已經到達
/* Check all fragment offsets to see if they connect. */
fp = qp->fragments;
offset = 0;
// 檢查所有分片，每個分片時按照偏移從小到大排序的鏈表，因為每次分片節點到達時會插入相應的位置
while (fp != NULL)
{ /*
如果當前節點的偏移大於期待的偏移(即上一個節點的最後一個位元組的偏移+1，由end欄位表示)，
說明有中間節點沒到達，直接返回未完成
*/
if (fp->offset > offset)
return(0); /* fragment(s) missing */
offset = fp->end;
fp = fp->next;
}

/* All fragments are present. */
// 分片全部到達並且每個分片的位元組連續則重組完成
return(1);
}
// 重組成功後構造完整的ip報文
static struct sk_buff *ip_glue(struct ipq *qp)
{
struct sk_buff *skb;
struct iphdr *iph;
struct ipfrag *fp;
unsigned char *ptr;
int count, len;

/*
* Allocate a new buffer for the datagram.
*/
// 整個包的長度等於mac頭長度+ip頭長度+數據長度
len = qp->maclen + qp->ihlen + qp->len;
// 分配新的skb
if ((skb = alloc_skb(len,GFP_ATOMIC)) == NULL)
{
ip_statistics.IpReasmFails++;
printk("IP: queue_glue: no memory for gluing queue 0x%X
", (int) qp);
ip_free(qp);
return(NULL);
}

/* Fill in the basic details. */
// 這裡應該是等於qp->len？
skb->len = (len - qp->maclen);
skb->h.raw = skb->data; // data欄位指向新分配的內存首地址
skb->free = 1;

/* Copy the original MAC and IP headers into the new buffer. */
ptr = (unsigned char *) skb->h.raw;
memcpy(ptr, ((unsigned char *) qp->mac), qp->maclen); // 把mac頭複製到新的內存
ptr += qp->maclen;
memcpy(ptr, ((unsigned char *) qp->iph), qp->ihlen); // 把ip頭複製到新的內存
ptr += qp->ihlen; // 指向數據部分的首地址
skb->h.raw += qp->maclen;// 指向ip頭首地址

count = 0;

/* Copy the data portions of all fragments into the new buffer. */
fp = qp->fragments;
// 開始複製數據部分
while(fp != NULL)
{ // 如果當前節點的數據長度+已經複製的內容長度大於skb->len則說明內容溢出了，丟棄該數據包
if(count+fp->len > skb->len)
{
printk("Invalid fragment list: Fragment over size.
");
ip_free(qp);
kfree_skb(skb,FREE_WRITE);
ip_statistics.IpReasmFails++;
return NULL;
}
// 把分片中的數據複製到對應偏移的位置
memcpy((ptr + fp->offset), fp->ptr, fp->len);
// 已複製的數據長度
count += fp->len;
fp = fp->next;
}

/* We glued together all fragments, so remove the queue entry. */
ip_free(qp);// 數據複製完後可以釋放分片隊列了

/* Done with all fragments. Fixup the new IP header. */
iph = skb->h.iph; // 上面的raw欄位指向了ip頭首地址，skb->h.iph等價於raw欄位的值
iph->frag_off = 0; // 清除分片欄位
// 更新總長度為ip頭+數據的長度
iph->tot_len = htons((iph->ihl * sizeof(unsigned long)) + count);
skb->ip_hdr = iph;

ip_statistics.IpReasmOKs++;
return(skb);
}
// 處理分片報文
static struct sk_buff *ip_defrag(struct iphdr *iph, struct sk_buff *skb, struct device *dev)
{
struct ipfrag *prev, *next;
struct ipfrag *tfp;
struct ipq *qp;
struct sk_buff *skb2;
unsigned char *ptr;
int flags, offset;
int i, ihl, end;

ip_statistics.IpReasmReqds++;

/* Find the entry of this IP datagram in the "incomplete datagrams" queue. */
qp = ip_find(iph); // 根據ip頭找是否已經存在分片隊列

/* Is this a non-fragmented datagram? */
offset = ntohs(iph->frag_off);
flags = offset & ~IP_OFFSET; // 取得三個分片標記位
offset &= IP_OFFSET; // 取得分片偏移
// 如果沒有更多分片了，並且offset=0（第一個分片），則屬於出錯，第一個分片後面肯定還有分片，否則幹嘛要分片
if (((flags & IP_MF) == 0) && (offset == 0))
{
if (qp != NULL)
ip_free(qp); /* Huh? How could this exist?? */
return(skb);
}
// 偏移乘以8得到數據的真實偏移
offset <<= 3; /* offset is in 8-byte chunks */

/*
* If the queue already existed, keep restarting its timer as long
* as we still are receiving fragments. Otherwise, create a fresh
* queue entry.
*/
/*
如果已經存在分片隊列，說明之前已經有分片到達，重置計時器，所以超時的邏輯是，
如果IP_FRAG_TIME時間內沒有分片到達，則認為重組超時，這裡沒有以總時間來判斷。
*/
if (qp != NULL)
{
del_timer(&qp->timer);
qp->timer.expires = IP_FRAG_TIME; /* about 30 seconds */
qp->timer.data = (unsigned long) qp; /* pointer to queue */
qp->timer.function = ip_expire; /* expire function */
add_timer(&qp->timer);
}
else
{
/*
* If we failed to create it, then discard the frame
*/
// 新建一個管理分片隊列的節點
if ((qp = ip_create(skb, iph, dev)) == NULL)
{
skb->sk = NULL;
kfree_skb(skb, FREE_READ);
ip_statistics.IpReasmFails++;
return NULL;
}
}

/*
* Determine the position of this fragment.
*/
// ip頭長度
ihl = (iph->ihl * sizeof(unsigned long));
// 偏移+數據部分長度等於end，end的值是最後一個位元組+1
end = offset + ntohs(iph->tot_len) - ihl;

/*
* Point into the IP datagram data part.
*/
// data指向整個報文首地址，即mac頭首地址，ptr指向ip報文的數據部分
ptr = skb->data + dev->hard_header_len + ihl;

/*
* Is this the final fragment?
*/
// 是否是最後一個分片，是的話，未分片的ip報文長度為end，即最後一個報文的最後一個位元組的偏移+1，因為偏移從0算起
if ((flags & IP_MF) == 0)
qp->len = end;

/*
* Find out which fragments are in front and at the back of us
* in the chain of fragments so far. We must know where to put
* this fragment, right?
*/

prev = NULL;
// 插入分片隊列相應的位置，保證分片的有序
for(next = qp->fragments; next != NULL; next = next->next)
{ // 找出第一個比當前分片偏移大的節點
if (next->offset > offset)
break; /* bingo! */
prev = next;
}

/*
* We found where to put this one.
* Check for overlap with preceding fragment, and, if needed,
* align things so that any overlaps are eliminated.
*/
// 處理分片重疊問題
/*
處理當前節點和前面節點的重疊問題，因為上面保證了offset >= prev->offset，
所以只需要比較當前節點的偏移和prev節點的end欄位
*/
if (prev != NULL && offset < prev->end)
{
// 說明存在重疊，算出重疊的大小，把當前節點的重疊部分丟棄，更新offset和ptr指針往前走,沒處理完全重疊的情況
i = prev->end - offset;
offset += i; /* ptr into datagram */
ptr += i; /* ptr into fragment data */
}

/*
* Look for overlap with succeeding segments.
* If we can merge fragments, do it.
*/
// 處理當前節點和後面節點的重疊問題
for(; next != NULL; next = tfp)
{
tfp = next->next;
// 當前節點及其後面的節點都不會發生重疊了
if (next->offset >= end)
break; /* no overlaps at all */
// 反之發生了重疊，算出重疊大小
i = end - next->offset; /* overlap is i bytes */
// 更新和當前節點重疊的節點的欄位，往後挪
next->len -= i; /* so reduce size of */
next->offset += i; /* next fragment */
next->ptr += i;

/*
* If we get a frag size of <= 0, remove it and the packet
* that it goes with.
*/
// 發生了完全重疊，則刪除舊的節點
if (next->len <= 0)
{
if (next->prev != NULL)
next->prev->next = next->next;// 說明舊節點不是第一個節點
else
qp->fragments = next->next;// 說明舊節點是第一個節點
// 這裡應該是tfp !=NULL ?
if (tfp->next != NULL)
next->next->prev = next->prev;

kfree_skb(next->skb,FREE_READ);
kfree_s(next, sizeof(struct ipfrag));
}
}

/*
* Insert this fragment in the chain of fragments.
*/

tfp = NULL;
// 創建一個分片節點
tfp = ip_frag_create(offset, end, skb, ptr);

/*
* No memory to save the fragment - so throw the lot
*/

if (!tfp)
{
skb->sk = NULL;
kfree_skb(skb, FREE_READ);
return NULL;
}
// 插入分片隊列
tfp->prev = prev;
tfp->next = next;
if (prev != NULL)
prev->next = tfp;
else
qp->fragments = tfp;

if (next != NULL)
next->prev = tfp;

/*
* OK, so we inserted this new fragment into the chain.
* Check if we now have a full IP datagram which we can
* bump up to the IP layer...
*/
// 判斷全部分片是否到達，是的話重組
if (ip_done(qp))
{
skb2 = ip_glue(qp); /* glue together the fragments */
return(skb2);
}
return(NULL);
}
// ip分片處理，即發出去的ip包太大需要分片
void ip_fragment(struct sock *sk, struct sk_buff *skb, struct device *dev, int is_frag)
{
struct iphdr *iph;
unsigned char *raw;
unsigned char *ptr;
struct sk_buff *skb2;
int left, mtu, hlen, len;
int offset;
unsigned long flags;

/*
* Point into the IP datagram header.
*/
// mac首地址
raw = skb->data;
// ip頭首地址
iph = (struct iphdr *) (raw + dev->hard_header_len);

skb->ip_hdr = iph;

/*
* Setup starting values.
*/

hlen = (iph->ihl * sizeof(unsigned long));
// 算出ip報文的數據長度
left = ntohs(iph->tot_len) - hlen; /* Space per frame */
hlen += dev->hard_header_len; /* Total header size */
// 每個分片的數據部分長度等於mac層的mtu減去mac頭和ip頭
mtu = (dev->mtu - hlen); /* Size of data space */
// 數據部分首地址
ptr = (raw + hlen); /* Where to start from */

/*
* Check for any "DF" flag. [DF means do not fragment]
*/
// 設置了不能分片則發送icmp報文
if (ntohs(iph->frag_off) & IP_DF)
{
/*
* Reply giving the MTU of the failed hop.
*/
ip_statistics.IpFragFails++;
icmp_send(skb,ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, dev->mtu, dev);
return;
}

/*
* The protocol doesnt seem to say what to do in the case that the
* frame + options doesnt fit the mtu. As it used to fall down dead
* in this case we were fortunate it didnt happen
*/
// mac頭的mtu小於8則直接返回，因為報文數據部分至少8個位元組
if(mtu<8)
{
/* Its wrong but its better than nothing */
icmp_send(skb,ICMP_DEST_UNREACH,ICMP_FRAG_NEEDED,dev->mtu, dev);
ip_statistics.IpFragFails++;
return;
}

// 該ip報文本身就是一個分片，現在需要進行再次分片，偏移的首地址是該報文的首地址乘以8
if (is_frag & 2)
offset = (ntohs(iph->frag_off) & 0x1fff) << 3;
else
offset = 0;

/*
* Keep copying data until we run out.
*/
// 開始分片
while(left > 0)
{
len = left;
/* IF: it doesnt fit, use mtu - the data space left */
// 大於mtu則繼續分片，否則就是最後一個分片
if (len > mtu)
len = mtu;
/* IF: we are not sending upto and including the packet end
then align the next start on an eight byte boundary */
// 剩下的位元組比mtu大的時候下面的判斷會成立，則取8的倍數大小，不一定等於mtu
if (len < left)
{
len/=8;
len*=8;
}
/*
* Allocate buffer.
*/
// 分片新的skb，大小為mac頭+ip頭+數據部分長度
if ((skb2 = alloc_skb(len + hlen,GFP_ATOMIC)) == NULL)
{
printk("IP: frag: no memory for new fragment!
");
ip_statistics.IpFragFails++;
return;
}

/*
* Set up data on packet
*/

skb2->arp = skb->arp;
if(skb->free==0)
printk("IP fragmenter: BUG free!=1 in fragmenter
");
skb2->free = 1;
skb2->len = len + hlen;
skb2->h.raw=(char *) skb2->data;
/*
* Charge the memory for the fragment to any owner
* it might possess
*/

save_flags(flags);
if (sk)
{
cli();
sk->wmem_alloc += skb2->mem_len;
skb2->sk=sk;
}
restore_flags(flags);
skb2->raddr = skb->raddr; /* For rebuild_header - must be here */

/*
* Copy the packet header into the new buffer.
*/
// 把mac報頭和ip報頭+選項都複製到skb中，選項應該只複製到第一個分片
memcpy(skb2->h.raw, raw, hlen);

/*
* Copy a block of the IP datagram.
*/
// 複製數據部分，長度為len，ptr指向原ip報文的首地址，
memcpy(skb2->h.raw + hlen, ptr, len);
left -= len;
// 指向ip頭首地址
skb2->h.raw+=dev->hard_header_len;

/*
* Fill in the new header fields.
*/
iph = (struct iphdr *)(skb2->h.raw/*+dev->hard_header_len*/);
// 設置該分配的偏移
iph->frag_off = htons((offset >> 3));
/*
* Added AC : If we are fragmenting a fragment thats not the
* last fragment then keep MF on each bit
*/
/*
1. 還有數據
2. 再分片的時候，該分片本身設置了分片flag，如果left大於MF置1，
如果left=0，需要看原報文是否設置了MF,如果有，說明原報文後面還有報文，
所以原報文下的所有分片MF都是1，如果原報文是最後一個報文，則MF=0，那對原報文分片的時候，
最後一個分片的MF=0，其他的為1
*/
if (left > 0 || (is_frag & 1))
iph->frag_off |= htons(IP_MF);
// 更新數據指針和偏移
ptr += len;
offset += len;

/*
* Put this fragment into the sending queue.
*/