From: "Zhang, Helin" <helin.zhang@intel.com>
To: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v3] mbuf: fix of enabling all newly added RX error flags
Date: Tue, 9 Dec 2014 02:23:39 +0000 [thread overview]
Message-ID: <F35DEAC7BCE34641BA9FAC6BCA4A12E70A7CFA7C@SHSMSX104.ccr.corp.intel.com> (raw)
In-Reply-To: <2601191342CEEE43887BDE71AB977258213BD9ED@IRSMSX105.ger.corp.intel.com>
> -----Original Message-----
> From: Ananyev, Konstantin
> Sent: Monday, December 8, 2014 6:45 PM
> To: Zhang, Helin; dev@dpdk.org
> Cc: Cao, Waterman; Cao, Min; olivier.matz@6wind.com
> Subject: RE: [PATCH v3] mbuf: fix of enabling all newly added RX error flags
>
> Hi Helin,
>
> > -----Original Message-----
> > From: Zhang, Helin
> > Sent: Saturday, December 06, 2014 1:34 AM
> > To: dev@dpdk.org
> > Cc: Cao, Waterman; Cao, Min; olivier.matz@6wind.com; Ananyev,
> > Konstantin; Zhang, Helin
> > Subject: [PATCH v3] mbuf: fix of enabling all newly added RX error
> > flags
> >
> > Before redefining mbuf structure, there was lack of free bits in 'ol_flags'
> > (32 bits in total) for new RX or TX flags. So it tried to reuse
> > existant bits as most as possible, or even assigning 0 to some of bit
> > flags. After new mbuf structure defined, there are quite a lot of free
> > bits. So those newly added bit flags should be assigned with correct
> > and valid bit values, and getting their names should be enabled as
> > well. Note that 'RECIP' should be removed, as nowhere will use it.
> > 'PKT_RX_ERR' is defined to replace all other error bit flags, e.g. MAC
> > error, Oversize error, header buffer overflow error.
> >
> > Signed-off-by: Helin Zhang <helin.zhang@intel.com>
> > ---
> > lib/librte_mbuf/rte_mbuf.c | 7 ++-----
> > lib/librte_mbuf/rte_mbuf.h | 7 ++-----
> > lib/librte_pmd_i40e/i40e_rxtx.c | 15 ++++-----------
> > 3 files changed, 8 insertions(+), 21 deletions(-)
> >
> > v2 changes:
> > * Removed error flag of 'ECIPE' processing only in both i40e PMD and mbuf.
> All
> > other error flags were added back.
> > * Assigned error flags with correct and valid values, as their previous values
> > were invalid.
> > * Enabled getting all error flag names.
> >
> > v3 changes:
> > * 'PKT_RX_ERR' is defined to replace error bit flags of MAC error, Oversize
> > error, header buffer overflow error.
> > * Removed assigning values to PKT_TX_* bit flags, as it has already been
> done
> > in another packet set.
> >
> > diff --git a/lib/librte_mbuf/rte_mbuf.c b/lib/librte_mbuf/rte_mbuf.c
> > index 1b14e02..5e6b5d0 100644
> > --- a/lib/librte_mbuf/rte_mbuf.c
> > +++ b/lib/librte_mbuf/rte_mbuf.c
> > @@ -210,11 +210,8 @@ const char *rte_get_rx_ol_flag_name(uint64_t
> mask)
> > case PKT_RX_FDIR: return "PKT_RX_FDIR";
> > case PKT_RX_L4_CKSUM_BAD: return "PKT_RX_L4_CKSUM_BAD";
> > case PKT_RX_IP_CKSUM_BAD: return "PKT_RX_IP_CKSUM_BAD";
> > - /* case PKT_RX_EIP_CKSUM_BAD: return "PKT_RX_EIP_CKSUM_BAD"; */
> > - /* case PKT_RX_OVERSIZE: return "PKT_RX_OVERSIZE"; */
> > - /* case PKT_RX_HBUF_OVERFLOW: return "PKT_RX_HBUF_OVERFLOW";
> */
> > - /* case PKT_RX_RECIP_ERR: return "PKT_RX_RECIP_ERR"; */
> > - /* case PKT_RX_MAC_ERR: return "PKT_RX_MAC_ERR"; */
> > + case PKT_RX_EIP_CKSUM_BAD: return "PKT_RX_EIP_CKSUM_BAD";
> > + case PKT_RX_ERR: return "PKT_RX_ERR";
> > case PKT_RX_IPV4_HDR: return "PKT_RX_IPV4_HDR";
> > case PKT_RX_IPV4_HDR_EXT: return "PKT_RX_IPV4_HDR_EXT";
> > case PKT_RX_IPV6_HDR: return "PKT_RX_IPV6_HDR"; diff --git
> > a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index
> > efdefc4..5e647a9 100644
> > --- a/lib/librte_mbuf/rte_mbuf.h
> > +++ b/lib/librte_mbuf/rte_mbuf.h
> > @@ -84,11 +84,6 @@ extern "C" {
> > #define PKT_RX_FDIR (1ULL << 2) /**< RX packet with FDIR
> match indicate. */
> > #define PKT_RX_L4_CKSUM_BAD (1ULL << 3) /**< L4 cksum of RX pkt.
> is
> > not OK. */ #define PKT_RX_IP_CKSUM_BAD (1ULL << 4) /**< IP cksum
> of
> > RX pkt. is not OK. */ -#define PKT_RX_EIP_CKSUM_BAD (0ULL << 0) /**<
> External IP header checksum error. */
> > -#define PKT_RX_OVERSIZE (0ULL << 0) /**< Num of desc of an RX
> pkt oversize. */
> > -#define PKT_RX_HBUF_OVERFLOW (0ULL << 0) /**< Header buffer
> overflow. */
> > -#define PKT_RX_RECIP_ERR (0ULL << 0) /**< Hardware processing
> error. */
> > -#define PKT_RX_MAC_ERR (0ULL << 0) /**< MAC error. */
> > #define PKT_RX_IPV4_HDR (1ULL << 5) /**< RX packet with IPv4
> header. */
> > #define PKT_RX_IPV4_HDR_EXT (1ULL << 6) /**< RX packet with
> extended IPv4 header. */
> > #define PKT_RX_IPV6_HDR (1ULL << 7) /**< RX packet with IPv6
> header. */
> > @@ -99,6 +94,8 @@ extern "C" {
> > #define PKT_RX_TUNNEL_IPV6_HDR (1ULL << 12) /**< RX tunnel packet
> with IPv6 header. */
> > #define PKT_RX_FDIR_ID (1ULL << 13) /**< FD id reported if FDIR
> match. */
> > #define PKT_RX_FDIR_FLX (1ULL << 14) /**< Flexible bytes reported
> if FDIR match. */
> > +#define PKT_RX_EIP_CKSUM_BAD (1ULL << 15) /**< External IP header
> checksum error. */
> > +#define PKT_RX_ERR (1ULL << 16) /**< Other errors, e.g.
> MAC error. */
> > /* add new RX flags here */
> >
> > /* add new TX flags here */
> > diff --git a/lib/librte_pmd_i40e/i40e_rxtx.c
> > b/lib/librte_pmd_i40e/i40e_rxtx.c index 2beae3c..5d99bd2 100644
> > --- a/lib/librte_pmd_i40e/i40e_rxtx.c
> > +++ b/lib/librte_pmd_i40e/i40e_rxtx.c
> > @@ -123,25 +123,18 @@ i40e_rxd_error_to_pkt_flags(uint64_t qword)
> > return flags;
> > /* If RXE bit set, all other status bits are meaningless */
> > if (unlikely(error_bits & (1 << I40E_RX_DESC_ERROR_RXE_SHIFT))) {
> > - flags |= PKT_RX_MAC_ERR;
> > + flags |= PKT_RX_ERR;
> > return flags;
> > }
> > -
> > - /* If RECIPE bit set, all other status indications should be ignored */
> > - if (unlikely(error_bits & (1 << I40E_RX_DESC_ERROR_RECIPE_SHIFT))) {
> > - flags |= PKT_RX_RECIP_ERR;
> > - return flags;
> > - }
> > - if (unlikely(error_bits & (1 << I40E_RX_DESC_ERROR_HBO_SHIFT)))
> > - flags |= PKT_RX_HBUF_OVERFLOW;
> > if (unlikely(error_bits & (1 << I40E_RX_DESC_ERROR_IPE_SHIFT)))
> > flags |= PKT_RX_IP_CKSUM_BAD;
> > if (unlikely(error_bits & (1 << I40E_RX_DESC_ERROR_L4E_SHIFT)))
> > flags |= PKT_RX_L4_CKSUM_BAD;
> > if (unlikely(error_bits & (1 << I40E_RX_DESC_ERROR_EIPE_SHIFT)))
> > flags |= PKT_RX_EIP_CKSUM_BAD;
> > - if (unlikely(error_bits & (1 << I40E_RX_DESC_ERROR_OVERSIZE_SHIFT)))
> > - flags |= PKT_RX_OVERSIZE;
> > + if (unlikely(error_bits & ((1 << I40E_RX_DESC_ERROR_OVERSIZE_SHIFT) ||
> > + (1 << I40E_RX_DESC_ERROR_HBO_SHIFT))))
> > + flags |= PKT_RX_ERR;
>
> It should be:
> error_bits & (1 << I40E_RX_DESC_ERROR_OVERSIZE_SHIFT | 1 <<
> I40E_RX_DESC_ERROR_HBO_SHIFT)
Ohm, stupid fault! I will fix it right now.
>
> Another 2 questions:
> - Why not to put all these 3 bits processing together:
> error_bits & (1 << I40E_RX_DESC_ERROR_OVERSIZE_SHIFT | 1 <<
> I40E_RX_DESC_ERROR_HBO_SHIFT | 1 <<
> I40E_RX_DESC_ERROR_RXE_SHIFT) ?
If RECIPE, other bits are meaningless, said by EAS.
If HBO or OVERSIZE, IPE, L4E, EIPE could be meaningful correct or not.
> - Is there any reason why you don't want to put PMD_RX_LOG(DEBUG, ...) for
> them?
> I think it might be usefull for debugging.
For all error bits or just HBO and OVERSIZE?
>
> Konstantin
>
>
> >
> > return flags;
> > }
> > --
> > 1.9.3
Regards,
Helin
next prev parent reply other threads:[~2014-12-09 2:23 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-26 6:07 [dpdk-dev] [PATCH] i40e: Use one bit flag for all hardware detected RX packet errors Helin Zhang
2014-11-26 10:49 ` Ananyev, Konstantin
2014-11-26 11:22 ` Olivier MATZ
2014-11-26 13:38 ` Ananyev, Konstantin
2014-11-26 14:12 ` Olivier MATZ
2014-11-28 8:07 ` Zhang, Helin
2014-11-28 8:47 ` Olivier MATZ
2014-12-01 1:57 ` Zhang, Helin
2014-12-01 9:58 ` Olivier MATZ
2014-12-02 7:25 ` Zhang, Helin
2014-12-05 1:46 ` [dpdk-dev] [PATCH v2 0/2] fix of enabling all newly added error flags Helin Zhang
2014-12-05 1:46 ` [dpdk-dev] [PATCH v2 1/2] i40e: remove checking rxd flag which is not public Helin Zhang
2014-12-05 1:46 ` [dpdk-dev] [PATCH v2 2/2] mbuf: assign valid bit values for some RX and TX flags Helin Zhang
2014-12-05 10:49 ` Ananyev, Konstantin
2014-12-06 0:42 ` Zhang, Helin
2014-12-06 1:07 ` Zhang, Helin
2014-12-08 10:55 ` Ananyev, Konstantin
2014-12-09 2:29 ` Zhang, Helin
2014-12-06 1:33 ` [dpdk-dev] [PATCH v3] mbuf: fix of enabling all newly added RX error flags Helin Zhang
2014-12-08 10:44 ` Ananyev, Konstantin
2014-12-09 2:23 ` Zhang, Helin [this message]
2014-12-08 10:50 ` Thomas Monjalon
2014-12-09 2:14 ` Zhang, Helin
2014-12-09 6:22 ` Thomas Monjalon
2014-12-10 8:55 ` [dpdk-dev] [PATCH v4] " Helin Zhang
2014-12-10 9:35 ` Thomas Monjalon
2014-12-10 13:50 ` Zhang, Helin
2014-12-10 15:26 ` Thomas Monjalon
2014-12-10 22:29 ` Zhang, Helin
2014-12-11 11:16 ` Olivier MATZ
2014-12-12 1:27 ` Zhang, Helin
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=F35DEAC7BCE34641BA9FAC6BCA4A12E70A7CFA7C@SHSMSX104.ccr.corp.intel.com \
--to=helin.zhang@intel.com \
--cc=dev@dpdk.org \
--cc=konstantin.ananyev@intel.com \
/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).