[dpdk-dev] [PATCH] Minor fixes in rte_common.h file.
Ravi Kerur
rkerur at gmail.com
Wed Dec 17 17:28:47 CET 2014
On Tue, Dec 16, 2014 at 5:05 PM, Ananyev, Konstantin <
konstantin.ananyev at intel.com> wrote:
>
>
>
> >
> > From: Ravi Kerur [mailto:rkerur at gmail.com]
> > Sent: Tuesday, December 16, 2014 8:14 PM
> > To: Ananyev, Konstantin
> > Cc: Neil Horman; dev at dpdk.org
> > Subject: Re: [dpdk-dev] [PATCH] Minor fixes in rte_common.h file.
> >
> >
> >
> > On Tue, Dec 16, 2014 at 9:23 AM, Ananyev, Konstantin <
> konstantin.ananyev at intel.com> wrote:
> >
> >
> > > -----Original Message-----
> > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ravi Kerur
> > > Sent: Tuesday, December 16, 2014 4:47 PM
> > > To: Neil Horman
> > > Cc: dev at dpdk.org
> > > Subject: Re: [dpdk-dev] [PATCH] Minor fixes in rte_common.h file.
> > >
> > > On Sat, Dec 13, 2014 at 2:39 AM, Neil Horman <nhorman at tuxdriver.com>
> wrote:
> > > >
> > > > On Fri, Dec 12, 2014 at 03:04:34PM -0800, r k wrote:
> > > > > Subject: [PATCH] Minor fixes in rte_common.h file.
> > > > >
> > > > > Fix rte_is_power_of_2 since 0 is not.
> > > > > Avoid branching instructions in RTE_MAX and RTE_MIN.
> > > > >
> > > > > Signed-off-by: Ravi Kerur <rkerur at gmail.com>
> > > > > ---
> > > > > lib/librte_eal/common/include/rte_common.h | 6 +++---
> > > > > lib/librte_pmd_e1000/igb_pf.c | 4 ++--
> > > > > lib/librte_pmd_ixgbe/ixgbe_pf.c | 4 ++--
> > > > > 3 files changed, 7 insertions(+), 7 deletions(-)
> > > > >
> > > > > diff --git a/lib/librte_eal/common/include/rte_common.h
> > > > > b/lib/librte_eal/common/include/rte_common.h
> > > > > index 921b91f..e163f35 100644
> > > > > --- a/lib/librte_eal/common/include/rte_common.h
> > > > > +++ b/lib/librte_eal/common/include/rte_common.h
> > > > > @@ -203,7 +203,7 @@ extern int RTE_BUILD_BUG_ON_detected_error;
> static
> > > > > inline int rte_is_power_of_2(uint32_t n) {
> > > > > - return ((n-1) & n) == 0;
> > > > > + return n && !(n & (n - 1));
> > > > > }
> > > > >
> > > > > /**
> > > > > @@ -259,7 +259,7 @@ rte_align64pow2(uint64_t v) #define
> RTE_MIN(a, b)
> > > > ({ \
> > > > > typeof (a) _a = (a); \
> > > > > typeof (b) _b = (b); \
> > > > > - _a < _b ? _a : _b; \
> > > > > + _b ^ ((_a ^ _b) & -(_a < _b)); \
> > > > Are you sure this is actually faster than the branch version? What
> about
> > > > using
> > > > a cmov instead?
> > > >
> > > >
> > > <rk> i am pretty sure modified code is faster than branching. I
> remember
> > > cmov had performance issues esp. on Pentuim-4 not sure how new intel
> cpu's
> > > perform.
> > I also think most modern compilers are smart enough to avoid any
> branching here and will use cmov instead.
> > And we are way ahead of Pentium 4 times these days.
> > Konstantin
> >
> > <rk>Konstantin, Can you please elaborate, is it something done
> automatically with Intel's icc compiler? My understanding is branch
> > prediction can be influenced only by using compiler builtin i.e.
> __builtin_expect() , without this compiler will generate regular
> > instructions(cmp/jump instructions). I wrote small program and compiled
> with gcc -02/-03, don't see cmov instruction.
>
> I am saying that there is probably no need to modify these macros.
> On IA , for constructions like: "_a < _b ? _a : _b;"
> modern compilers in many cases will avoid any branches and emit cmov
> instead.
>
> $ cat tcmv1.c
>
> #include <stdint.h>
> #include <stddef.h>
>
> #define RTE_MIN(a, b) ({ \
> typeof (a) _a = (a); \
> typeof (b) _b = (b); \
> _a < _b ? _a : _b; \
> })
>
> int
> fxmini32(int a, int b)
> {
> return RTE_MIN(a, b);
> }
>
> int
> fxminu64(uint64_t a, uint64_t b)
> {
> return RTE_MIN(a, b);
> }
>
> $gcc -O3 -m64 -S tcmv1.c
>
> $ cat tcmv1.s
> .file "tcmv1.c"
> .text
> .p2align 4,,15
> .globl fxmini32
> .type fxmini32, @function
> fxmini32:
> .LFB0:
> .cfi_startproc
> cmpl %esi, %edi
> movl %esi, %eax
> cmovle %edi, %eax
> ret
> .cfi_endproc
> .LFE0:
> .size fxmini32, .-fxmini32
> .p2align 4,,15
> .globl fxminu64
> .type fxminu64, @function
> fxminu64:
> .LFB1:
> .cfi_startproc
> cmpq %rsi, %rdi
> movq %rsi, %rax
> cmovbe %rdi, %rax
> ret
> .cfi_endproc
>
> gcc version 4.8.3
> clang produces similar code.
>
> Konstantin
>
<rk> Thanks, looks like it depends on gcc version. I have
.ident "GCC: (GNU) 4.4.6 20110731 (Red Hat 4.4.6-3)"
.section .note.GNU-stack,"", at progbits
...
and it still generates cmp/jump instructions. Anyways i will drop this
patch and just send fix for power_of_2.
-Ravi
> >
> > >
> > > > })
> > > > >
> > > > > /**
> > > > > @@ -268,7 +268,7 @@ rte_align64pow2(uint64_t v) #define
> RTE_MAX(a, b)
> > > > ({ \
> > > > > typeof (a) _a = (a); \
> > > > > typeof (b) _b = (b); \
> > > > > - _a > _b ? _a : _b; \
> > > > > + _a ^ ((_a ^ _b) & -(_a < _b)); \
> > > > Same as above
> > > >
> > > > <rk> Same as above.
> > >
> > > > > })
> > > > >
> > > > > /*********** Other general functions / macros ********/ diff --git
> > > > > a/lib/librte_pmd_e1000/igb_pf.c b/lib/librte_pmd_e1000/igb_pf.c
> index
> > > > > bc3816a..546499c 100644
> > > > > --- a/lib/librte_pmd_e1000/igb_pf.c
> > > > > +++ b/lib/librte_pmd_e1000/igb_pf.c
> > > > > @@ -321,11 +321,11 @@ igb_vf_set_mac_addr(struct rte_eth_dev *dev,
> > > > uint32_t
> > > > > vf, uint32_t *msgbuf) static int igb_vf_set_multicast(struct
> > > > rte_eth_dev
> > > > > *dev, __rte_unused uint32_t vf, uint32_t *msgbuf) {
> > > > > - int i;
> > > > > + int16_t i;
> > > > > uint32_t vector_bit;
> > > > > uint32_t vector_reg;
> > > > > uint32_t mta_reg;
> > > > > - int entries = (msgbuf[0] & E1000_VT_MSGINFO_MASK) >>
> > > > > + int32_t entries = (msgbuf[0] & E1000_VT_MSGINFO_MASK) >>
> > > > > E1000_VT_MSGINFO_SHIFT;
> > > > NAK, this has nothing to do with the included changelog
> > > >
> > >
> > > <rk> It does, it causes compilation errors such as
> > >
> > > /root/dpdk-new/dpdk/lib/librte_pmd_e1000/igb_pf.c: In function
> > > \u2018igb_pf_mbx_process\u2019:
> > > /root/dpdk-new/dpdk/lib/librte_pmd_e1000/igb_pf.c:350:23: error: array
> > > subscript is above array bounds [-Werror=array-bounds]
> > > vfinfo->vf_mc_hashes[i] = hash_list[i];
> > > ^
> > > cc1: all warnings being treated as errors
> > >
> > > Also it is always better to use explicit int definitions esp. for 64bit
> > > systems.
> > >
> > >
> > >
> > > >
> > > > > uint16_t *hash_list = (uint16_t *)&msgbuf[1];
> > > > > struct e1000_hw *hw =
> > > > > E1000_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> > > > > diff --git a/lib/librte_pmd_ixgbe/ixgbe_pf.c
> > > > > b/lib/librte_pmd_ixgbe/ixgbe_pf.c index 51da1fd..426caf9 100644
> > > > > --- a/lib/librte_pmd_ixgbe/ixgbe_pf.c
> > > > > +++ b/lib/librte_pmd_ixgbe/ixgbe_pf.c
> > > > > @@ -390,7 +390,7 @@ ixgbe_vf_set_multicast(struct rte_eth_dev *dev,
> > > > > __rte_unused uint32_t vf, uint32
> > > > > struct ixgbe_hw *hw =
> > > > > IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
> > > > > struct ixgbe_vf_info *vfinfo =
> > > > >
> *(IXGBE_DEV_PRIVATE_TO_P_VFDATA(dev->data->dev_private));
> > > > > - int nb_entries = (msgbuf[0] & IXGBE_VT_MSGINFO_MASK) >>
> > > > > + int32_t nb_entries = (msgbuf[0] & IXGBE_VT_MSGINFO_MASK) >>
> > > > > IXGBE_VT_MSGINFO_SHIFT;
> > > > ditto
> > > > > uint16_t *hash_list = (uint16_t *)&msgbuf[1];
> > > > > uint32_t mta_idx;
> > > > > @@ -399,7 +399,7 @@ ixgbe_vf_set_multicast(struct rte_eth_dev *dev,
> > > > > __rte_unused uint32_t vf, uint32
> > > > > const uint32_t IXGBE_MTA_BIT_SHIFT = 5;
> > > > > const uint32_t IXGBE_MTA_BIT_MASK = (0x1 <<
> IXGBE_MTA_BIT_SHIFT)
> > > > -
> > > > > 1;
> > > > > uint32_t reg_val;
> > > > > - int i;
> > > > > + int16_t i;
> > > > ditto
> > > >
> > > > <rk> Same as above.
> > >
> > > > >
> > > > > /* only so many hash values supported */
> > > > > nb_entries = RTE_MIN(nb_entries, IXGBE_MAX_VF_MC_ENTRIES);
> > > > > --
> > > > > 1.9.1
> > > > >
> > > >
>
More information about the dev
mailing list