[dpdk-dev] [PATCH 5/7] pci: support multiple PCI regions per device

2013-06-05 Thread Damien Millescamps
On 06/05/2013 05:49 PM, Stephen Hemminger wrote: > On Wed, 05 Jun 2013 16:50:03 +0200 > Damien Millescamps wrote: > >> Hi Stephen, >> >> Overall this patch is very nice. My only comment on this one is why do >> you limit the max number of memory resources to 5 ? >> The PCI configuration space perm

[dpdk-dev] [PATCH 5/7] pci: support multiple PCI regions per device

2013-06-05 Thread Damien Millescamps
Hi Stephen, Overall this patch is very nice. My only comment on this one is why do you limit the max number of memory resources to 5 ? The PCI configuration space permits to store up to 6 base addresses. > +#define PCI_MEM_RESOURCE 5 Please, can you add a log/comment with your patch, too ? Che

[dpdk-dev] [PATCH 7/7] eal: add ability to override DPDK syslog parameters

2013-06-05 Thread Vincent JARDIN
On 30/05/2013 19:12, Stephen Hemminger wrote: > By default, DPDK based applications would only allow logging > to syslog as "rte", DAEMON; but for any production application more > control is desired to allow using actual application name and > overriding the facility. > > Signed-off-by: Stephen He

[dpdk-dev] [PATCH 3/7] optimize log/panic

2013-06-05 Thread Vincent JARDIN
On 30/05/2013 19:12, Stephen Hemminger wrote: > Both logging and calls to panic are never in the critical path. > Use the GCC attribute cold to mark these functions as cold, > which generates more optimised code. > > Signed-off-by: Stephen Hemminger Reviewed-by: Vincent Jardin It does not hurt to

[dpdk-dev] [PATCH 2/7] rte_timer: optimize for empty case

2013-06-05 Thread Vincent JARDIN
On 30/05/2013 19:12, Stephen Hemminger wrote: > In many application there are no timers queued, and the call to > rte_timer_managecan be optimized in that case avoid reading HPET and > lock overhead. > > Signed-off-by: Stephen Hemminger Reviewed-by: Vincent Jardin It is a nice to have for perform

[dpdk-dev] [PATCH 1/7] [PATCH 4/8] igb: workaround errata with wthresh on 82576

2013-06-05 Thread Vincent JARDIN
On 30/05/2013 19:12, Stephen Hemminger wrote: > The 82576 has known issues which require the write threshold to be set to 1. > See: > http://download.intel.com/design/network/specupdt/82576_SPECUPDATE.pdf > > If not then single packets will hang in transmit ring until more arrive. > Simple te

[dpdk-dev] [RFC] igb_uio: rework interrupt using pci_intx_mask

2013-06-05 Thread Stephen Hemminger
The code to do interrupts in the igb_uio is broken on later kernel versions because pci_cfg_access_lock must not be used from interrupt context. But there is a much better way of doing all this now that a generic access to INTX is available. Use pci_intx routines in later kernels, and provide back

[dpdk-dev] [PATCH 5/7] pci: support multiple PCI regions per device

2013-06-05 Thread Stephen Hemminger
On Wed, 05 Jun 2013 20:05:15 +0200 Damien Millescamps wrote: > On 06/05/2013 05:49 PM, Stephen Hemminger wrote: > > On Wed, 05 Jun 2013 16:50:03 +0200 > > Damien Millescamps wrote: > > > >> Hi Stephen, > >> > >> Overall this patch is very nice. My only comment on this one is why do > >> you limi

[dpdk-dev] [PATCH 5/7] pci: support multiple PCI regions per device

2013-06-05 Thread Stephen Hemminger
On Wed, 05 Jun 2013 16:50:03 +0200 Damien Millescamps wrote: > Hi Stephen, > > Overall this patch is very nice. My only comment on this one is why do > you limit the max number of memory resources to 5 ? > The PCI configuration space permits to store up to 6 base addresses. > > > +#define PCI_M