Commit Graph

87 Commits

Author SHA1 Message Date
Konstantin Belousov
a2622e5dc2 Restore the segment registers and segment base MSRs for amd64 syscall
return path only when neither thread was context switched while
executing syscall code nor syscall explicitely modified LDT or MSRs.

Save segment registers in trap handlers before interrupts are enabled,
to not allow context switches to happen before registers are saved.
Use separated byte in pcb for indication of fast/full return, since
pcb_flags are not synchronized with context switches.

The change puts back syscall microbenchmark numbers that were slowed
down after commit of the support for LDT on amd64.

Reviewed by:	jeff
Tested (and tested, and tested ...) by:	pho
Approved by:	re (kensmith)
2009-07-09 09:34:11 +00:00
Konstantin Belousov
2c66cccab7 Save and restore segment registers on amd64 when entering and leaving
the kernel on amd64. Fill and read segment registers for mcontext and
signals. Handle traps caused by restoration of the
invalidated selectors.

Implement user-mode creation and manipulation of the process-specific
LDT descriptors for amd64, see sysarch(2).

Implement support for TSS i/o port access permission bitmap for amd64.

Context-switch LDT and TSS. Do not save and restore segment registers on
the context switch, that is handled by kernel enter/leave trampolines
now. Remove segment restore code from the signal trampolines for
freebsd/amd64, freebsd/ia32 and linux/i386 for the same reason.

Implement amd64-specific compat shims for sysarch.

Linuxolator (temporary ?) switched to use gsbase for thread_area pointer.

TODO:
Currently, gdb is not adapted to show segment registers from struct reg.
Also, no machine-depended ptrace command is added to set segment
registers for debugged process.

In collaboration with:	pho
Discussed with:	peter
Reviewed by:	jhb
Linuxolator tested by:	dchagin
2009-04-01 13:09:26 +00:00
John Baldwin
2ee8325f42 A better fix for handling different FPU initial control words for different
ABIs:
- Store the FPU initial control word in the pcb for each thread.
- When first using the FPU, load the initial control word after restoring
  the clean state if it is not the standard control word.
- Provide a correct control word for Linux/i386 binaries under
  FreeBSD/amd64.
- Adjust the control word returned for fpugetregs()/npxgetregs() when a
  thread hasn't used the FPU yet to reflect the real initial control
  word for the current ABI.
- The Linux/i386 ABI for FreeBSD/i386 now properly sets the right control
  word instead of trashing whatever the current state of the FPU is.

Reviewed by:	bde
2009-03-05 19:42:11 +00:00
David E. O'Brien
e6493bbebf Change some movl's to mov's. Newer GAS no longer accept 'movl' instructions
for moving between a segment register and a 32-bit memory location.

Looked at by:	jhb
2009-01-31 11:37:21 +00:00
Konstantin Belousov
f98c3ea74e - When executing FreeBSD/amd64 binaries from FreeBSD/i386 or Linux/i386
processes, clear PCB_32BIT and PCB_GS32BIT bits [1].

- Reread the fs and gs bases from the msr unconditionally, not believing
  the values in pcb_fsbase and pcb_gsbase, since usermode may reload
  segment registers, invalidating the cache. [2].

Both problems resulted in the wrong fs base, causing wrong tls pointer
be dereferenced in the usermode.

Reported and tested by:	Vyacheslav Bocharov <adeepv at gmail com> [1]
Reported by:	Bernd Walter <ticsoat cicely7 cicely de>,
	Artem Belevich <fbsdlist at src cx>[2]
Reviewed by:	peter
MFC after:	3 days
2008-09-02 17:52:11 +00:00
Konstantin Belousov
8f4a1f3a83 Bring back the save/restore of the %ds, %es, %fs and %gs registers for
the 32bit images on amd64.

Change the semantic of the PCB_32BIT pcb flag to request the context
switch code to operate on the segment registers. Its previous meaning
of saving or restoring the %gs base offset is assigned to the new
PCB_GS32BIT flag.

FreeBSD 32bit image activator sets the PCB_32BIT flag, while Linux 32bit
emulation sets PCB_32BIT | PCB_GS32BIT.

Reviewed by:	peter
MFC after:	2 weeks
2008-07-30 11:30:55 +00:00
Peter Wemm
3d60169ef4 Protect the setting of the fsbase/gsbase MSR registers and the
pcb_[fg]sbase values with a critical section, like the rest of the kernel.
2008-03-23 22:44:56 +00:00
Konstantin Belousov
22eca0bf45 Since version 4.3, gcc changed its behaviour concerning the i386/amd64
ABI and the direction flag, that is it now assumes that the direction
flag is cleared at the entry of a function and it doesn't clear once
more if needed. This new behaviour conforms to the i386/amd64 ABI.

Modify the signal handler frame setup code to clear the DF {e,r}flags
bit on the amd64/i386 for the signal handlers.

jhb@ noted that it might break old apps if they assumed DF == 1 would be
preserved in the signal handlers, but that such apps should be rare and
that older versions of gcc would not generate such apps.

Submitted by:	Aurelien Jarno <aurelien aurel32 net>
PR:	121422
Reviewed by:	jhb
MFC after:	2 weeks
2008-03-13 10:54:38 +00:00
Attilio Rao
393a081d42 Optimize vmmeter locking.
In particular:
- Add an explicative table for locking of struct vmmeter members
- Apply new rules for some of those members
- Remove some unuseful comments

Heavily reviewed by: alc, bde, jeff
Approved by: jeff (mentor)
2007-06-10 21:59:14 +00:00
Attilio Rao
6759608248 Rework the PCPU_* (MD) interface:
- Rename PCPU_LAZY_INC into PCPU_INC
- Add the PCPU_ADD interface which just does an add on the pcpu member
  given a specific value.

Note that for most architectures PCPU_INC and PCPU_ADD are not safe.
This is a point that needs some discussions/work in the next days.

Reviewed by: alc, bde
Approved by: jeff (mentor)
2007-06-04 21:38:48 +00:00
Kip Macy
e5f8d4099d Newer versions of gcc don't support treating structures passed by value
as if they were really passed by reference. Specifically, the dead stores
elimination pass in the GCC 4.1 optimiser breaks the non-compliant behavior
on which FreeBSD relied. This change brings FreeBSD up to date by switching
trap frames to being explicitly passed by reference.

Reviewed by: kan
Tested by: kan
2006-12-17 06:48:40 +00:00
David Xu
c6511aea86 Move some declaration of 32-bit signal structures into file
freebsd32-signal.h, implement sigtimedwait and sigwaitinfo system calls.
2006-10-05 01:56:11 +00:00
David Xu
07a8ebcc75 Stop reloading %fs and %gs, since it causes the base address from
GDT to be loaded into FS.base and GS.base, these values of course
are not the values set by sysarch() with I386_SET_FSBASE and
I386_SET_GSBASE, the change fixed a crash for 32bit libthr after
signal handler returned and normal code is accessing thread pointer,
for example: movl %gs:8, %eax.
2006-09-23 13:42:09 +00:00
John Baldwin
cb76d9b05c Retire SYF_ARGMASK and remove both SYF_MPSAFE and SYF_ARGMASK. sy_narg is
now back to just being an argument count.
2006-07-28 20:22:58 +00:00
John Baldwin
af5bf12239 Now that all system calls are MPSAFE, retire the SYF_MPSAFE flag used to
mark system calls as being MPSAFE:
- Stop conditionally acquiring Giant around system call invocations.
- Remove all of the 'M' prefixes from the master system call files.
- Remove support for the 'M' prefix from the script that generates the
  syscall-related files from the master system call files.
- Don't explicitly set SYF_MPSAFE when registering nfssvc.
2006-07-28 19:05:28 +00:00
John Baldwin
22ea1bc57a Unify the checking for lock misbehavior in the various syscall()
implementations and adjust some of the checks while I'm here:
- Add a new check to make sure we don't return from a syscall in a critical
  section.
- Add a new explicit check before userret() to make sure we don't return
  with any locks held.  The advantage here is that we can include the
  syscall number and name in syscall() whereas that info is not available
  in userret().
- Drop the mtx_assert()'s of sched_lock and Giant.  They are replaced by
  the more general checks just added.

MFC after:	2 weeks
2006-07-27 22:32:30 +00:00
John Baldwin
0c5d1dbd43 Add KTR_SYSC tracing to the syscall() implementations that didn't have it
yet.

MFC after:	1 week
2006-07-27 21:25:50 +00:00
John Baldwin
00f1856905 Add missing ptrace(2) system-call stops to various syscall()
implementations.

MFC after:	1 week
2006-07-27 19:50:16 +00:00
Poul-Henning Kamp
eb2da9a51f Simplify system time accounting for profiling.
Rename struct thread's td_sticks to td_pticks, we will need the
other name for more appropriately named use shortly.  Reduce it
from uint64_t to u_int.

Clear td_pticks whenever we enter the kernel instead of recording
its value as reference for userret().  Use the absolute value of
td->pticks in userret() and eliminate third argument.
2006-02-08 08:09:17 +00:00
Wayne Salamon
4f9ac41fba Call the audit syscall enter/exit functions for the amd64 architecture,
both 32-bit and 64-bit paths. System calls will now be audited.

Obtained from: TrustedBSD Project
Approved by: rwatson (mentor)
2006-02-04 20:37:20 +00:00
Peter Wemm
79880f7327 Catch up to the system siginfo changes. Use a union for the ia32 layout
of siginfo just like the system one.  There are now two fields to copy
instead of one.
2005-12-06 23:06:29 +00:00
Ruslan Ermilov
160ebe8754 Catch up with the recent <sys/signal.h> change and make this compile. 2005-11-04 20:32:26 +00:00
Paul Saab
bbf719c8ba include opt_compat.h to unbreak the build 2005-10-24 00:00:00 +00:00
David Xu
9104847f21 1. Change prototype of trapsignal and sendsig to use ksiginfo_t *, most
changes in MD code are trivial, before this change, trapsignal and
   sendsig use discrete parameters, now they uses member fields of
   ksiginfo_t structure. For sendsig, this change allows us to pass
   POSIX realtime signal value to user code.

2. Remove cpu_thread_siginfo, it is no longer needed because we now always
   generate ksiginfo_t data and feed it to libpthread.

3. Add p_sigqueue to proc structure to hold shared signals which were
   blocked by all threads in the proc.

4. Add td_sigqueue to thread structure to hold all signals delivered to
   thread.

5. i386 and amd64 now return POSIX standard si_code, other arches will
   be fixed.

6. In this sigqueue implementation, pending signal set is kept as before,
   an extra siginfo list holds additional siginfo_t data for signals.
   kernel code uses psignal() still behavior as before, it won't be failed
   even under memory pressure, only exception is when deleting a signal,
   we should call sigqueue_delete to remove signal from sigqueue but
   not SIGDELSET. Current there is no kernel code will deliver a signal
   with additional data, so kernel should be as stable as before,
   a ksiginfo can carry more information, for example, allow signal to
   be delivered but throw away siginfo data if memory is not enough.
   SIGKILL and SIGSTOP have fast path in sigqueue_add, because they can
   not be caught or masked.
   The sigqueue() syscall allows user code to queue a signal to target
   process, if resource is unavailable, EAGAIN will be returned as
   specification said.
   Just before thread exits, signal queue memory will be freed by
   sigqueue_flush.
   Current, all signals are allowed to be queued, not only realtime signals.

Earlier patch reviewed by: jhb, deischen
Tested on: i386, amd64
2005-10-14 12:43:47 +00:00
Peter Wemm
07f5921b86 Don't set segment registers via ptrace yet. Its not ready. 2005-10-04 23:26:00 +00:00
Peter Wemm
add121a476 Implement 32 bit getcontext/setcontext/swapcontext on amd64. I've added
stubs for ia64 to keep it compiling.  These are used by 32 bit apps such
as gdb.
2005-09-27 18:04:20 +00:00
Peter Wemm
62919d788b Jumbo-commit to enhance 32 bit application support on 64 bit kernels.
This is good enough to be able to run a RELENG_4 gdb binary against
a RELENG_4 application, along with various other tools (eg: 4.x gcore).
We use this at work.

ia32_reg.[ch]: handle the 32 bit register file format, used by ptrace,
	procfs and core dumps.
procfs_*regs.c: vary the format of proc/XXX/*regs depending on the client
	and target application.
procfs_map.c: Don't print a 64 bit value to 32 bit consumers, or their
	sscanf fails.  They expect an unsigned long.
imgact_elf.c: produce a valid 32 bit coredump for 32 bit apps.
sys_process.c: handle 32 bit consumers debugging 32 bit targets.  Note
	that 64 bit consumers can still debug 32 bit targets.

IA64 has got stubs for ia32_reg.c.

Known limitations: a 5.x/6.x gdb uses get/setcontext(), which isn't
implemented in the 32/64 wrapper yet.  We also make a tiny patch to
gdb pacify it over conflicting formats of ld-elf.so.1.

Approved by:	re
2005-06-30 07:49:22 +00:00
John Baldwin
aa9aa68d2f Use PCPU_LAZY_INC() for cnt.v_{intr,trap,syscalls} rather than atomic
operations in some places and simple non-per CPU math in others.
2005-04-12 23:18:54 +00:00
Julian Elischer
99e9dcb817 Remove sched_free_thread() which was only used
in diagnostics. It has outlived its usefulness and has started
causing panics for people who turn on DIAGNOSTIC, in what is otherwise
good code.

MFC after:	2 days
2004-08-31 06:12:13 +00:00
Bruce Evans
909ca1671d Don't repeat the definition of IDTVEC(). It is in asmacros.h. 2004-05-24 11:28:11 +00:00
Bruce Evans
7a9e253666 Added profiling support for Xint0x80_syscall. 2004-05-23 19:06:15 +00:00
Warner Losh
9a80fddc71 Remove advertising clause from University of California Regent's license,
per letter dated July 22, 1999 and email from Peter Wemm.

Approved by: core, peter
2004-04-05 23:55:14 +00:00
Peter Wemm
4fa84e0fb3 Catch up with some proc/procsig locking improvements that were made to the
i386 version and were not merged over.
2004-02-21 23:34:42 +00:00
David Xu
a30ec4b99c Make sigaltstack as per-threaded, because per-process sigaltstack state
is useless for threaded programs, multiple threads can not share same
stack.
The alternative signal stack is private for thread, no lock is needed,
the orignal P_ALTSTACK is now moved into td_pflags and renamed to
TDP_ALTSTACK.
For single thread or Linux clone() based threaded program, there is no
semantic changed, because those programs only have one kernel thread
in every process.

Reviewed by: deischen, dfr
2004-01-03 02:02:26 +00:00
Peter Wemm
3554ecf032 Catch up with the procsig locking changes elsewhere. We were doing
things like copyin/out with psp->ps_mtx held.  That was not good.

Approved by:  re (scottl)
2003-12-03 07:00:30 +00:00
Peter Wemm
0d2a298904 Initial landing of SMP support for FreeBSD/amd64.
- This is heavily derived from John Baldwin's apic/pci cleanup on i386.
- I have completely rewritten or drastically cleaned up some other parts.
  (in particular, bootstrap)
- This is still a WIP.  It seems that there are some highly bogus bioses
  on nVidia nForce3-150 boards.  I can't stress how broken these boards
  are.  I have a workaround in mind, but right now the Asus SK8N is broken.
  The Gigabyte K8NPro (nVidia based) is also mind-numbingly hosed.
- Most of my testing has been with SCHED_ULE.  SCHED_4BSD works.
- the apic and acpi components are 'standard'.
- If you have an nVidia nForce3-150 board, you are stuck with 'device
  atpic' in addition, because they somehow managed to forget to connect the
  8254 timer to the apic, even though its in the same silicon!  ARGH!
  This directly violates the ACPI spec.
2003-11-17 08:58:16 +00:00
Peter Wemm
e46fe241d3 Stop pretending to support kernel profiling. The FAKE_MCOUNT() etc
calls are just gradually getting more and more stale.  At this point it
would be better to start from scratch once prof_machdep.c is adapted.
2003-11-13 02:38:33 +00:00
Peter Wemm
4cd2d525e3 Move a MD 32 bit binary support routine into the MD areas. exec_setregs
is highly MD in an emulation environment since it operates on the host
environment.  Although the setregs functions are really for exec support
rather than signals, they deal with the same sorts of context and include
files.  So I put it there rather than create yet another file.
2003-11-08 07:43:44 +00:00
Peter Wemm
bf2f09ee97 The great s/npx/fpu/gi 2003-11-08 03:33:38 +00:00
Peter Wemm
76caec589f MFi386 by jhb: use symbolic constants for the IDT entries. 2003-09-22 22:09:02 +00:00
Peter Wemm
46159d1fd6 Switch to using the emulator in the common compat area.
Still work-in-progress.
2003-08-23 00:04:53 +00:00
Peter Wemm
c639ca93f4 Initial sweep at dividing up the generic 32bit-on-64bit kernel support
from the ia32 specific stuff.  Some of this still needs to move to the MI
freebsd32 area, and some needs to move to the MD area.  This is still
work-in-progress.
2003-08-22 23:19:02 +00:00
Peter Wemm
82914097e5 Regen 2003-08-21 03:48:50 +00:00
Peter Wemm
6b59055cb8 This is too funny for words. Swap syscalls 416 and 417 around. It works
better that way when sigaction() and sigreturn() do the right thing.
2003-08-21 03:48:05 +00:00
David E. O'Brien
56ae44c5df Use __FBSDID().
Brought to you by:	a boring talk at Ottawa Linux Symposium
2003-07-25 21:19:19 +00:00
Peter Wemm
5681a6f60d Regenerate. 2003-05-31 06:51:04 +00:00
Peter Wemm
1f5b79bc16 Make this compile with WITNESS enabled. It wants the syscall names. 2003-05-31 06:49:53 +00:00
Peter Wemm
9f0c4ab393 Deal with the user VM space expanding. 32 bit applications do not like
having their stack at the 512GB mark.  Give 4GB of user VM space for 32
bit apps.  Note that this is significantly more than on i386 which gives
only about 2.9GB of user VM to a process (1GB for kernel, plus page
table pages which eat user VM space).

Approved by: re (blanket)
2003-05-23 05:07:33 +00:00
Peter Wemm
c0a54ff621 Collect the nastiness for preserving the kernel MSR_GSBASE around the
load_gs() calls into a single place that is less likely to go wrong.

Eliminate the per-process context switching of MSR_GSBASE, because it
should be constant for a single cpu.  Instead, save/restore it during
the loading of the new %gs selector for the new process.

Approved by:	re (amd64/* blanket)
2003-05-15 00:23:40 +00:00
Peter Wemm
e14528b349 Regen
Approved by: re (amd64 blanket)
2003-05-14 04:11:25 +00:00