building on Mac
Kazu Yamamoto ( 山本和彦 )
kazu at iij.ad.jp
Thu Mar 14 07:18:43 CET 2013
Hi,
First of all, I would like to ask whether or not someone can build
(not "validate") GHC head on Mountain Lion. The reason why I ask is
that I found this issue:
http://stackoverflow.com/questions/13539066/can-write-to-a-non-blocking-fd-return-eagain-when-select-reports-it-as-writabl
And I would like to ask what about on Lion or on Snow Leopard? Is this
building failure specific to Mountain Lion?
Here is what I know so far:
- This issue happens when
libraries/template-haskell/Language/Haskell/TH/Syntax.hs is
compiled.
- According to "dtruss", "select()" called by ghc-stage1 is waiting a
pipe forever.
- No "gcc" is executed during this compiling.
- This happens with both 32bit bootstrapping GHC and 64bit one.
Attached is a log of "dtruss".
Another question: Is ghc-stage1 compiled with "-threaded"?
--Kazu
69117/0x10d80b2: pipe(0x7FFF6DDBF240, 0x0, 0x0) = 3 0
69117/0x10d80b2: pipe(0x7FFF6DDBF230, 0x0, 0x4) = 5 0
69117/0x10d80b2: pipe(0x7FFF6DDBF220, 0x0, 0x6) = 7 0
69117/0x10d80b2: sigprocmask(0x1, 0x110248EA8, 0x110248EAC) = 0x0 0
69117/0x10d80b2: setitimer(0x0, 0x7FFF6DDBF1B0, 0x0) = 0 0
69149/0x10d80b2: vfork() = 69149 0
69149/0x10d80b2: sigprocmask(0x3, 0x110248EAC, 0x0) = 0x0 0
69149/0x10d80b2: dup2(0x3, 0x0, 0x0) = 0 0
69149/0x10d80b2: close(0x3) = 0 0
69149/0x10d80b2: close(0x4) = 0 0
69149/0x10d80b2: dup2(0x6, 0x1, 0x0) = 1 0
69149/0x10d80b2: close(0x6) = 0 0
69149/0x10d80b2: close(0x5) = 0 0
69149/0x10d80b2: dup2(0x8, 0x2, 0x0) = 2 0
69149/0x10d80b2: close(0x8) = 0 0
69149/0x10d80b2: close(0x7) = 0 0
69117/0x10d80b2: execve("/usr/bin/gcc\0", 0x113F943E0, 0x7FFF6DDC35C0) = 69149 0
69117/0x10d80b2: close(0x3) = 0 0
69117/0x10d80b2: fcntl(0x4, 0x2, 0x1) = 0 0
69117/0x10d80b2: close(0x6) = 0 0
69117/0x10d80b2: fcntl(0x5, 0x2, 0x1) = 0 0
69117/0x10d80b2: close(0x8) = 0 0
69117/0x10d80b2: fcntl(0x7, 0x2, 0x1) = 0 0
69117/0x10d80b2: sigprocmask(0x3, 0x110248EAC, 0x0) = 0x0 0
69117/0x10d80b2: setitimer(0x0, 0x7FFF6DDBF1B0, 0x0) = 0 0
69117/0x10d80b2: fcntl(0x4, 0x3, 0x11E00A40) = 1 0
69117/0x10d80b2: fcntl(0x4, 0x4, 0x5) = 0 0
69117/0x10d80b2: ioctl(0x4, 0x4004667A, 0x7Fdtrace: error on enabled probe ID 1791 (ID 116: syscall::read:return): out of scratch space in action #13 at DIF offset 44
dtrace: error on enabled probe ID 1791 (ID 116: syscall::read:return): out of scratch space in action #13 at DIF offset 44
FF6DDBF294) = -1 Err#25
69117/0x10d80b2: ioctl(0x4, 0x40487413, 0x7FFF6DDBF248) = -1 Err#25
69117/0x10d80b2: fcntl(0x5, 0x3, 0x11E00A40) = 0 0
69117/0x10d80b2: fcntl(0x5, 0x4, 0x4) = 0 0
69117/0x10d80b2: ioctl(0x5, 0x4004667A, 0x7FFF6DDBF294) = -1 Err#25
69117/0x10d80b2: ioctl(0x5, 0x40487413, 0x7FFF6DDBF248) = -1 Err#25
69117/0x10d80b2: fcntl(0x7, 0x3, 0x11E00A40) = 0 0
69117/0x10d80b2: fcntl(0x7, 0x4, 0x4) = 0 0
69117/0x10d80b2: ioctl(0x7, 0x4004667A, 0x7FFF6DDBF294) = -1 Err#25
69117/0x10d80b2: ioctl(0x7, 0x40487413, 0x7FFF6DDBF248) = -1 Err#25
69117/0x10d80b2: ioctl(0x5, 0x4004667A, 0x7FFF6DDBF294) = -1 Err#25
69117/0x10d80b2: ioctl(0x5, 0x40487413, 0x7FFF6DDBF248) = -1 Err#25
69117/0x10d80b2: ioctl(0x7, 0x4004667A, 0x7FFF6DDBF294) = -1 Err#25
69117/0x10d80b2: ioctl(0x7, 0x40487413, 0x7FFF6DDBF248) = -1 Err#25
69117/0x10d80b2: wait4(0x10E1D, 0x7FFF6DDBF28C, 0x1) = 0 0
69117/0x10d80b2: select(0x6, 0x7FFF6DDC3220, 0x7FFF6DDC31A0, 0x0, 0x7FFF6DDC32A0) = 0 0
69117/0x10d80b2: setitimer(0x0, 0x7FFF6DDC3140, 0x0) = 0 0
More information about the ghc-devs
mailing list