From: Jason Wang <jasowang@redhat.com>
To: qemu-devel@nongnu.org, peter.maydell@linaro.org
Cc: Akihiko Odaki <akihiko.odaki@daynix.com>,
Sriram Yagnaraman <sriram.yagnaraman@est.tech>,
Jason Wang <jasowang@redhat.com>
Subject: [PULL 10/50] igb: Always copy ethernet header
Date: Tue, 23 May 2023 15:31:58 +0800 [thread overview]
Message-ID: <20230523073238.54236-11-jasowang@redhat.com> (raw)
In-Reply-To: <20230523073238.54236-1-jasowang@redhat.com>
From: Akihiko Odaki <akihiko.odaki@daynix.com>
igb_receive_internal() used to check the iov length to determine
copy the iovs to a contiguous buffer, but the check is flawed in two
ways:
- It does not ensure that iovcnt > 0.
- It does not take virtio-net header into consideration.
The size of this copy is just 22 octets, which can be even less than
the code size required for checks. This (wrong) optimization is probably
not worth so just remove it. Removing this also allows igb to assume
aligned accesses for the ethernet header.
Fixes: 3a977deebe ("Intrdocue igb device emulation")
Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Reviewed-by: Sriram Yagnaraman <sriram.yagnaraman@est.tech>
Signed-off-by: Jason Wang <jasowang@redhat.com>
---
hw/net/igb_core.c | 43 +++++++++++++++++++++++--------------------
1 file changed, 23 insertions(+), 20 deletions(-)
diff --git a/hw/net/igb_core.c b/hw/net/igb_core.c
index 21a8d9a..1123df9 100644
--- a/hw/net/igb_core.c
+++ b/hw/net/igb_core.c
@@ -67,6 +67,11 @@ typedef struct IGBTxPktVmdqCallbackContext {
NetClientState *nc;
} IGBTxPktVmdqCallbackContext;
+typedef struct L2Header {
+ struct eth_header eth;
+ struct vlan_header vlan;
+} L2Header;
+
static ssize_t
igb_receive_internal(IGBCore *core, const struct iovec *iov, int iovcnt,
bool has_vnet, bool *external_tx);
@@ -961,15 +966,16 @@ igb_rx_is_oversized(IGBCore *core, uint16_t qn, size_t size)
return size > (lpe ? max_ethernet_lpe_size : max_ethernet_vlan_size);
}
-static uint16_t igb_receive_assign(IGBCore *core, const struct eth_header *ehdr,
+static uint16_t igb_receive_assign(IGBCore *core, const L2Header *l2_header,
size_t size, E1000E_RSSInfo *rss_info,
bool *external_tx)
{
static const int ta_shift[] = { 4, 3, 2, 0 };
+ const struct eth_header *ehdr = &l2_header->eth;
uint32_t f, ra[2], *macp, rctl = core->mac[RCTL];
uint16_t queues = 0;
uint16_t oversized = 0;
- uint16_t vid = lduw_be_p(&PKT_GET_VLAN_HDR(ehdr)->h_tci) & VLAN_VID_MASK;
+ uint16_t vid = be16_to_cpu(l2_header->vlan.h_tci) & VLAN_VID_MASK;
bool accepted = false;
int i;
@@ -1590,14 +1596,13 @@ static ssize_t
igb_receive_internal(IGBCore *core, const struct iovec *iov, int iovcnt,
bool has_vnet, bool *external_tx)
{
- static const int maximum_ethernet_hdr_len = (ETH_HLEN + 4);
-
uint16_t queues = 0;
uint32_t n = 0;
- uint8_t min_buf[ETH_ZLEN];
+ union {
+ L2Header l2_header;
+ uint8_t octets[ETH_ZLEN];
+ } buf;
struct iovec min_iov;
- struct eth_header *ehdr;
- uint8_t *filter_buf;
size_t size, orig_size;
size_t iov_ofs = 0;
E1000E_RxRing rxr;
@@ -1623,24 +1628,21 @@ igb_receive_internal(IGBCore *core, const struct iovec *iov, int iovcnt,
net_rx_pkt_unset_vhdr(core->rx_pkt);
}
- filter_buf = iov->iov_base + iov_ofs;
orig_size = iov_size(iov, iovcnt);
size = orig_size - iov_ofs;
/* Pad to minimum Ethernet frame length */
- if (size < sizeof(min_buf)) {
- iov_to_buf(iov, iovcnt, iov_ofs, min_buf, size);
- memset(&min_buf[size], 0, sizeof(min_buf) - size);
+ if (size < sizeof(buf)) {
+ iov_to_buf(iov, iovcnt, iov_ofs, &buf, size);
+ memset(&buf.octets[size], 0, sizeof(buf) - size);
e1000x_inc_reg_if_not_full(core->mac, RUC);
- min_iov.iov_base = filter_buf = min_buf;
- min_iov.iov_len = size = sizeof(min_buf);
+ min_iov.iov_base = &buf;
+ min_iov.iov_len = size = sizeof(buf);
iovcnt = 1;
iov = &min_iov;
iov_ofs = 0;
- } else if (iov->iov_len < maximum_ethernet_hdr_len) {
- /* This is very unlikely, but may happen. */
- iov_to_buf(iov, iovcnt, iov_ofs, min_buf, maximum_ethernet_hdr_len);
- filter_buf = min_buf;
+ } else {
+ iov_to_buf(iov, iovcnt, iov_ofs, &buf, sizeof(buf.l2_header));
}
/* Discard oversized packets if !LPE and !SBP. */
@@ -1648,11 +1650,12 @@ igb_receive_internal(IGBCore *core, const struct iovec *iov, int iovcnt,
return orig_size;
}
- ehdr = PKT_GET_ETH_HDR(filter_buf);
- net_rx_pkt_set_packet_type(core->rx_pkt, get_eth_packet_type(ehdr));
+ net_rx_pkt_set_packet_type(core->rx_pkt,
+ get_eth_packet_type(&buf.l2_header.eth));
net_rx_pkt_set_protocols(core->rx_pkt, iov, iovcnt, iov_ofs);
- queues = igb_receive_assign(core, ehdr, size, &rss_info, external_tx);
+ queues = igb_receive_assign(core, &buf.l2_header, size,
+ &rss_info, external_tx);
if (!queues) {
trace_e1000e_rx_flt_dropped();
return orig_size;
--
2.7.4
next prev parent reply other threads:[~2023-05-23 7:33 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-23 7:31 [PULL 00/50] Net patches Jason Wang
2023-05-23 7:31 ` [PULL 01/50] e1000e: Fix tx/rx counters Jason Wang
2023-05-23 7:31 ` [PULL 02/50] hw/net/net_tx_pkt: Decouple implementation from PCI Jason Wang
2023-05-23 7:31 ` [PULL 03/50] hw/net/net_tx_pkt: Decouple interface " Jason Wang
2023-05-23 7:31 ` [PULL 04/50] e1000x: Fix BPRC and MPRC Jason Wang
2023-05-23 7:31 ` [PULL 05/50] igb: Fix Rx packet type encoding Jason Wang
2023-05-23 7:31 ` [PULL 06/50] igb: Do not require CTRL.VME for tx VLAN tagging Jason Wang
2023-05-23 7:31 ` [PULL 07/50] igb: Clear IMS bits when committing ICR access Jason Wang
2023-05-23 7:31 ` [PULL 08/50] net/net_rx_pkt: Use iovec for net_rx_pkt_set_protocols() Jason Wang
2023-05-23 7:31 ` [PULL 09/50] e1000e: Always copy ethernet header Jason Wang
2023-05-23 7:31 ` Jason Wang [this message]
2023-05-23 7:31 ` [PULL 11/50] Fix references to igb Avocado test Jason Wang
2023-05-23 7:32 ` [PULL 12/50] tests/avocado: Remove unused imports Jason Wang
2023-05-23 7:32 ` [PULL 13/50] tests/avocado: Remove test_igb_nomsi_kvm Jason Wang
2023-05-23 7:32 ` [PULL 14/50] hw/net/net_tx_pkt: Remove net_rx_pkt_get_l4_info Jason Wang
2023-05-23 7:32 ` [PULL 15/50] net/eth: Rename eth_setup_vlan_headers_ex Jason Wang
2023-05-23 7:32 ` [PULL 16/50] e1000x: Share more Rx filtering logic Jason Wang
2023-05-23 7:32 ` [PULL 17/50] e1000x: Take CRC into consideration for size check Jason Wang
2023-05-23 7:32 ` [PULL 18/50] e1000x: Rename TcpIpv6 into TcpIpv6Ex Jason Wang
2023-05-23 7:32 ` [PULL 19/50] e1000e: Always log status after building rx metadata Jason Wang
2023-05-23 7:32 ` [PULL 20/50] igb: " Jason Wang
2023-05-23 7:32 ` [PULL 21/50] igb: Remove goto Jason Wang
2023-05-23 7:32 ` [PULL 22/50] igb: Read DCMD.VLE of the first Tx descriptor Jason Wang
2023-05-23 7:32 ` [PULL 23/50] e1000e: Reset packet state after emptying Tx queue Jason Wang
2023-05-23 7:32 ` [PULL 24/50] vmxnet3: " Jason Wang
2023-05-23 7:32 ` [PULL 25/50] igb: Add more definitions for Tx descriptor Jason Wang
2023-05-23 7:32 ` [PULL 26/50] igb: Share common VF constants Jason Wang
2023-05-23 7:32 ` [PULL 27/50] igb: Fix igb_mac_reg_init coding style alignment Jason Wang
2023-05-23 7:32 ` [PULL 28/50] igb: Clear EICR bits for delayed MSI-X interrupts Jason Wang
2023-05-23 7:32 ` [PULL 29/50] e1000e: Rename a variable in e1000e_receive_internal() Jason Wang
2023-05-23 7:32 ` [PULL 30/50] igb: Rename a variable in igb_receive_internal() Jason Wang
2023-05-23 7:32 ` [PULL 31/50] net/eth: Use void pointers Jason Wang
2023-05-23 7:32 ` [PULL 32/50] net/eth: Always add VLAN tag Jason Wang
2023-05-23 7:32 ` [PULL 33/50] hw/net/net_rx_pkt: Enforce alignment for eth_header Jason Wang
2023-05-23 7:32 ` [PULL 34/50] tests/qtest/libqos/igb: Set GPIE.Multiple_MSIX Jason Wang
2023-05-23 7:32 ` [PULL 35/50] igb: Implement MSI-X single vector mode Jason Wang
2023-05-23 7:32 ` [PULL 36/50] igb: Use UDP for RSS hash Jason Wang
2023-05-23 7:32 ` [PULL 37/50] igb: Implement Rx SCTP CSO Jason Wang
2023-05-23 7:32 ` [PULL 38/50] igb: Implement Tx " Jason Wang
2023-05-23 7:32 ` [PULL 39/50] igb: Strip the second VLAN tag for extended VLAN Jason Wang
2023-05-23 7:32 ` [PULL 40/50] igb: Filter with " Jason Wang
2023-05-23 7:32 ` [PULL 41/50] igb: Implement igb-specific oversize check Jason Wang
2023-05-23 7:32 ` [PULL 42/50] igb: Implement Rx PTP2 timestamp Jason Wang
2023-05-23 7:32 ` [PULL 43/50] igb: Implement Tx timestamp Jason Wang
2023-05-23 7:32 ` [PULL 44/50] e1000e: Notify only new interrupts Jason Wang
2023-05-23 7:32 ` [PULL 45/50] igb: " Jason Wang
2023-05-23 7:32 ` [PULL 46/50] igb: Clear-on-read ICR when ICR.INTA is set Jason Wang
2023-05-23 7:32 ` [PULL 47/50] vmxnet3: Do not depend on PC Jason Wang
2023-05-23 7:32 ` [PULL 48/50] MAINTAINERS: Add a reviewer for network packet abstractions Jason Wang
2023-05-23 7:32 ` [PULL 49/50] docs/system/devices/igb: Note igb is tested for DPDK Jason Wang
2023-05-23 7:32 ` [PULL 50/50] rtl8139: fix large_send_mss divide-by-zero Jason Wang
2023-05-23 17:56 ` [PULL 00/50] Net patches Richard Henderson
2023-05-23 19:53 ` Michael Tokarev
2023-05-24 4:06 ` Jason Wang
2023-05-24 4:21 ` Akihiko Odaki
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=20230523073238.54236-11-jasowang@redhat.com \
--to=jasowang@redhat.com \
--cc=akihiko.odaki@daynix.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=sriram.yagnaraman@est.tech \
/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).