February 2011 - kernelnewbies@kernelnewbies.org

BUILD_BUG_ON_ZERO
by Zhang Meng Feb. 26, 2011

Feb. 26, 2011

Hi ~List, Could anybody explain the macro below? what does it mean? Thanks in advance. #define BUILD_BUG_ON_ZERO(e) (sizeof(struct { int:-!!(e); })) -- Yours sincerely ZhangMeng

3 2

Re: syscalls performance
by Daniel Baluta Feb. 25, 2011

Feb. 25, 2011

On Sat, Feb 26, 2011 at 12:03 AM, Mauro Romano Trajber <trajber(a)gmail.com> wrote: > Same behavior for a new syscall created from scratch. Instruction caching?

2 1

Re: syscalls performance
by Daniel Baluta Feb. 25, 2011

Feb. 25, 2011

On Fri, Feb 25, 2011 at 8:22 PM, Mauro Romano Trajber <trajber(a)gmail.com> wrote: > Thanks Enrico and Daniel, you're right. glibc was caching getpid(); but this > is not the root cause of this behavior. > Going further, I decide to use call getpid without glibc, using > syscall(SYS_getpid) to test this behavior and it happened again. > Calling it once, the test consumes about 7k CPU cycles and 10 calls consumes > about 10k CPU cycles. > Any ideas ? Can you post a pointer to your code and information about how you got this numbers? thanks, Daniel.

2 1

syscalls performance
by Mauro Romano Trajber Feb. 25, 2011

Feb. 25, 2011

I was doing some performance tests of system calls and I find an interesting behavior. Using RDTSC to count the CPU cycles, a single call to the getpid() consumes about 7k of CPU clock cycles and ten calls consume approximately 9,800 cycles. The fact is that from the second call, the CPU cycles grows at a rate of about 350 CPU cycles per call. Why does this happen? There is some hardware optimization when the syscall ID is already in EAX register ? Any ideias ?

2 2

Re: syscalls performance
by Enrico Granata Feb. 25, 2011

Feb. 25, 2011

If you are calling getpid() as a libc function, the library is caching the result. This means, after the first time calling it (when you really have to go into the kernel), subsequent calls will simply return the cached value without ever leaving user level Enrico Granata Computer Science & Engineering Department (EBU3B) - Room 3240 office phone 858 534 9914 University of California, San Diego > On Feb 25, 2011, at 9:03 AM, Mauro Romano Trajber wrote: > >> I was doing some performance tests of system calls and I find an interesting behavior. >> Using RDTSC to count the CPU cycles, a single call to the getpid() consumes about 7k of CPU clock cycles and ten calls consume approximately 9,800 cycles. >> The fact is that from the second call, the CPU cycles grows at a rate of about 350 CPU cycles per call. >> Why does this happen? There is some hardware optimization when the syscall ID is already in EAX register ? >> >> Any ideias ? >> _______________________________________________ >> Kernelnewbies mailing list >> Kernelnewbies(a)kernelnewbies.org >> http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies >

1 0

PCI express, link down problem..
by SHANKARCHEKURI Feb. 25, 2011

Feb. 25, 2011

We have designed a SBC with MPC8548 PCIe 4x lanes connected with IDT PCIe swicth.My POR of processor is showing the correct configuration as 4x with 2.5Gbps,RC mode.But when i read the PCIe LTSSM register it shows lane is in Detect.quiet.when i probed the actual signals Processor side -Tx diff.voltage of one lane(Lane 0) shows 700mv pp and it is reaching the Rx.Remaining all other lane voltages are 0mv.Similarly,Receiver side -TX,only one lane diff.voltage is showing 112mv pp,remainig all other lanes are showing 0mv.what is the expected voltage levels?.The negotiated link width is showing 1x.Is it due to the same reason?what could be reason for the detect.quiet state? -- View this message in context: http://old.nabble.com/PCI-express%2C-link-down-problem..-tp31010155p3101015… Sent from the kernelnewbies mailing list archive at Nabble.com.

1 0

UML related query
by mohit verma Feb. 24, 2011

Feb. 24, 2011

hi all if i put a break point on a function in UML and want to pass arguments to it , how can it be done? thanks in advance. -- ........................ *MOHIT VERMA*

2 1

Fwd: UML related query
by mohit verma Feb. 24, 2011

Feb. 24, 2011

hi all if i put a break point on a function in UML and want to pass arguments to it , how can it be done? ( i am putting the breakpoint on a kernel function) thanks in advance. -- ........................ *MOHIT VERMA* -- ........................ *MOHIT VERMA*

1 0

Macro definitions for FL_DATA and FL_IMAGE
by Madhavi Manchala Feb. 24, 2011

Feb. 24, 2011

Dear All, In one of the makefile souces, I saw the following statement. arm-elf-gcc util.c -o flutil -DFL_DATA=util_bin -DFL_IMAGE=util.bin -DFL_DATA_CKSUM=$FL_DATA_CKSUM I do not understand about the FL_DATA and FL_IMAGE macros that are passed in the above statement. Where can I get the information on these macros? Any idea / links, would be appreciated. Thanks and Regards, Madhavi M.

2 3

Debugging memset crashes
by Praveen kumar Feb. 24, 2011

Feb. 24, 2011

Hi, I have an embedded system on which i get crash (Caused by memset ) at long run( Reproducible 2/10 times ), I wanted to know efficient ways (materials) to handle memset crashes . *Program terminated with signal 11, Segmentation fault. * #0 memset (dstpp=0xd2, c=<value optimized out>, len=3374) at ../../../src/Common/OS/linux/memset.c:20 20 ../../../src/Common/OS/linux/memset.c: No such file or directory. in ../../../src/Common/OS/linux/memset.c - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - *(1) Display the backtrace for the thread that crashed. * #0 memset (dstpp=0xd2, c=<value optimized out>, len=3374) at ../../../src/Common/OS/linux/memset.c:20 #1 0x100e76c4 in LPREncode (pEnc=0x5b1debc8, pRTPPacket=0x5be69ee4) at ../../../src/Common/comm/lpr/lpr.c:647 #2 0x100dde50 in RTPChanWork (pChan=<value optimized out>) at ../../../src/Common/comm/rtpchan/rtpchan.c:2699 #3 0x100de5d8 in RTPTxTask (userDefined=<value optimized out>) at ../../../src/Common/comm/rtpchan/rtpchan.c:2853 #4 0x1051d744 in pthread_entry (args=<value optimized out>) at ../../../src/Common/OS/linux/psosapi/pthreadapi.c:222 #5 0x0ffaadec in start_thread (arg=<value optimized out>) at pthread_create.c:298 #6 0x0f892530 in clone () from /lib/libc.so.6 Previous frame inner to this frame (corrupt stack?) - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - *(2) Display argument parameters and local variables for the thread. * dstpp = (void *) 0xd2 c = <value optimized out> len = 3374 xlen = <value optimized out> cccc = <value optimized out> dstp = 210 do anyone have some information.. Thanks in Advance Praveen

2 1