3 Light-weight System Calls for IA-64
4 -----------------------------------
7 Last update: 27-Sep-2003
12 Using the "epc" instruction effectively introduces a new mode of
13 execution to the ia64 linux kernel. We call this mode the
14 "fsys-mode". To recap, the normal states of execution are:
17 Both the register stack and the memory stack have been
18 switched over to kernel memory. The user-level state is saved
19 in a pt-regs structure at the top of the kernel memory stack.
22 Both the register stack and the kernel stack are in
23 user memory. The user-level state is contained in the
26 - bank 0 interruption-handling mode:
27 This is the non-interruptible state which all
28 interruption-handlers start execution in. The user-level
29 state remains in the CPU registers and some kernel state may
30 be stored in bank 0 of registers r16-r31.
32 In contrast, fsys-mode has the following special properties:
34 - execution is at privilege level 0 (most-privileged)
36 - CPU registers may contain a mixture of user-level and kernel-level
37 state (it is the responsibility of the kernel to ensure that no
38 security-sensitive kernel-level state is leaked back to
41 - execution is interruptible and preemptible (an fsys-mode handler
42 can disable interrupts and avoid all other interruption-sources
45 - neither the memory-stack nor the register-stack can be trusted while
46 in fsys-mode (they point to the user-level stacks, which may
47 be invalid, or completely bogus addresses)
49 In summary, fsys-mode is much more similar to running in user-mode
50 than it is to running in kernel-mode. Of course, given that the
51 privilege level is at level 0, this means that fsys-mode requires some
55 * How to tell fsys-mode
57 Linux operates in fsys-mode when (a) the privilege level is 0 (most
58 privileged) and (b) the stacks have NOT been switched to kernel memory
59 yet. For convenience, the header file <asm-ia64/ptrace.h> provides
66 The "regs" argument is a pointer to a pt_regs structure. The "task"
67 argument is a pointer to the task structure to which the "regs"
68 pointer belongs to. user_mode() returns TRUE if the CPU state pointed
69 to by "regs" was executing in user mode (privilege level 3).
70 user_stack() returns TRUE if the state pointed to by "regs" was
71 executing on the user-level stack(s). Finally, fsys_mode() returns
72 TRUE if the CPU state pointed to by "regs" was executing in fsys-mode.
73 The fsys_mode() macro is equivalent to the expression:
75 !user_mode(regs) && user_stack(task,regs)
77 * How to write an fsyscall handler
79 The file arch/ia64/kernel/fsys.S contains a table of fsyscall-handlers
80 (fsyscall_table). This table contains one entry for each system call.
81 By default, a system call is handled by fsys_fallback_syscall(). This
82 routine takes care of entering (full) kernel mode and calling the
83 normal Linux system call handler. For performance-critical system
84 calls, it is possible to write a hand-tuned fsyscall_handler. For
85 example, fsys.S contains fsys_getpid(), which is a hand-tuned version
86 of the getpid() system call.
88 The entry and exit-state of an fsyscall handler is as follows:
90 ** Machine state on entry to fsyscall handler:
93 - r11 = saved ar.pfs (a user-level value)
94 - r15 = system call number
95 - r16 = "current" task pointer (in normal kernel-mode, this is in r13)
96 - r32-r39 = system call arguments
97 - b6 = return address (a user-level value)
98 - ar.pfs = previous frame-state (a user-level value)
99 - PSR.be = cleared to zero (i.e., little-endian byte order is in effect)
100 - all other registers may contain values passed in from user-mode
102 ** Required machine state on exit to fsyscall handler:
104 - r11 = saved ar.pfs (as passed into the fsyscall handler)
105 - r15 = system call number (as passed into the fsyscall handler)
106 - r32-r39 = system call arguments (as passed into the fsyscall handler)
107 - b6 = return address (as passed into the fsyscall handler)
108 - ar.pfs = previous frame-state (as passed into the fsyscall handler)
110 Fsyscall handlers can execute with very little overhead, but with that
111 speed comes a set of restrictions:
113 o Fsyscall-handlers MUST check for any pending work in the flags
114 member of the thread-info structure and if any of the
115 TIF_ALLWORK_MASK flags are set, the handler needs to fall back on
116 doing a full system call (by calling fsys_fallback_syscall).
118 o Fsyscall-handlers MUST preserve incoming arguments (r32-r39, r11,
119 r15, b6, and ar.pfs) because they will be needed in case of a
120 system call restart. Of course, all "preserved" registers also
121 must be preserved, in accordance to the normal calling conventions.
123 o Fsyscall-handlers MUST check argument registers for containing a
124 NaT value before using them in any way that could trigger a
125 NaT-consumption fault. If a system call argument is found to
126 contain a NaT value, an fsyscall-handler may return immediately
127 with r8=EINVAL, r10=-1.
129 o Fsyscall-handlers MUST NOT use the "alloc" instruction or perform
130 any other operation that would trigger mandatory RSE
131 (register-stack engine) traffic.
133 o Fsyscall-handlers MUST NOT write to any stacked registers because
134 it is not safe to assume that user-level called a handler with the
135 proper number of arguments.
137 o Fsyscall-handlers need to be careful when accessing per-CPU variables:
138 unless proper safe-guards are taken (e.g., interruptions are avoided),
139 execution may be pre-empted and resumed on another CPU at any given
142 o Fsyscall-handlers must be careful not to leak sensitive kernel'
143 information back to user-level. In particular, before returning to
144 user-level, care needs to be taken to clear any scratch registers
145 that could contain sensitive information (note that the current
146 task pointer is not considered sensitive: it's already exposed
149 o Fsyscall-handlers MUST NOT access user-memory without first
150 validating access-permission (this can be done typically via
151 probe.r.fault and/or probe.w.fault) and without guarding against
152 memory access exceptions (this can be done with the EX() macros
153 defined by asmmacro.h).
155 The above restrictions may seem draconian, but remember that it's
156 possible to trade off some of the restrictions by paying a slightly
157 higher overhead. For example, if an fsyscall-handler could benefit
158 from the shadow register bank, it could temporarily disable PSR.i and
159 PSR.ic, switch to bank 0 (bsw.0) and then use the shadow registers as
160 needed. In other words, following the above rules yields extremely
161 fast system call execution (while fully preserving system call
162 semantics), but there is also a lot of flexibility in handling more
167 The delivery of (asynchronous) signals must be delayed until fsys-mode
168 is exited. This is accomplished with the help of the lower-privilege
169 transfer trap: arch/ia64/kernel/process.c:do_notify_resume_user()
170 checks whether the interrupted task was in fsys-mode and, if so, sets
171 PSR.lp and returns immediately. When fsys-mode is exited via the
172 "br.ret" instruction that lowers the privilege level, a trap will
173 occur. The trap handler clears PSR.lp again and returns immediately.
174 The kernel exit path then checks for and delivers any pending signals.
178 The "epc" instruction doesn't change the contents of PSR at all. This
179 is in contrast to a regular interruption, which clears almost all
180 bits. Because of that, some care needs to be taken to ensure things
181 work as expected. The following discussion describes how each PSR bit
184 PSR.be Cleared when entering fsys-mode. A srlz.d instruction is used
185 to ensure the CPU is in little-endian mode before the first
186 load/store instruction is executed. PSR.be is normally NOT
187 restored upon return from an fsys-mode handler. In other
188 words, user-level code must not rely on PSR.be being preserved
189 across a system call.
192 PSR.mfl Unchanged. Note: fsys-mode handlers must not write-registers!
193 PSR.mfh Unchanged. Note: fsys-mode handlers must not write-registers!
194 PSR.ic Unchanged. Note: fsys-mode handlers can clear the bit, if needed.
195 PSR.i Unchanged. Note: fsys-mode handlers can clear the bit, if needed.
198 PSR.dfl Unchanged. Note: fsys-mode handlers must not write-registers!
199 PSR.dfh Unchanged. Note: fsys-mode handlers must not write-registers!
204 PSR.db Unchanged. The kernel prevents user-level from setting a hardware
205 breakpoint that triggers at any privilege level other than 3 (user-mode).
207 PSR.tb Lazy redirect. If a taken-branch trap occurs while in
208 fsys-mode, the trap-handler modifies the saved machine state
209 such that execution resumes in the gate page at
210 syscall_via_break(), with privilege level 3. Note: the
211 taken branch would occur on the branch invoking the
212 fsyscall-handler, at which point, by definition, a syscall
213 restart is still safe. If the system call number is invalid,
214 the fsys-mode handler will return directly to user-level. This
215 return will trigger a taken-branch trap, but since the trap is
216 taken _after_ restoring the privilege level, the CPU has already
217 left fsys-mode, so no special treatment is needed.
219 PSR.cpl Cleared to 0.
220 PSR.is Unchanged (guaranteed to be 0 on entry to the gate page).
222 PSR.it Unchanged (guaranteed to be 1).
223 PSR.id Unchanged. Note: the ia64 linux kernel never sets this bit.
224 PSR.da Unchanged. Note: the ia64 linux kernel never sets this bit.
225 PSR.dd Unchanged. Note: the ia64 linux kernel never sets this bit.
226 PSR.ss Lazy redirect. If set, "epc" will cause a Single Step Trap to
227 be taken. The trap handler then modifies the saved machine
228 state such that execution resumes in the gate page at
229 syscall_via_break(), with privilege level 3.
231 PSR.ed Unchanged. Note: This bit could only have an effect if an fsys-mode
232 handler performed a speculative load that gets NaTted. If so, this
233 would be the normal & expected behavior, so no special treatment is
235 PSR.bn Unchanged. Note: fsys-mode handlers may clear the bit, if needed.
236 Doing so requires clearing PSR.i and PSR.ic as well.
237 PSR.ia Unchanged. Note: the ia64 linux kernel never sets this bit.
239 * Using fast system calls
241 To use fast system calls, userspace applications need simply call
242 __kernel_syscall_via_epc(). For example
244 -- example fgettimeofday() call --
245 -- fgettimeofday.S --
247 #include <asm/asmmacro.h>
249 GLOBAL_ENTRY(fgettimeofday)
255 mov r2 = 0xa000000000020660;; // gate address
256 // found by inspection of System.map for the
257 // __kernel_syscall_via_epc() function. See
258 // below for how to do this for real.
261 mov r15 = 1087 // gettimeofday syscall
263 br.call.sptk.many b6 = b7
269 br.ret.sptk.many rp;; // return to caller
272 -- end fgettimeofday.S --
274 In reality, getting the gate address is accomplished by two extra
275 values passed via the ELF auxiliary vector (include/asm-ia64/elf.h)
277 o AT_SYSINFO : is the address of __kernel_syscall_via_epc()
278 o AT_SYSINFO_EHDR : is the address of the kernel gate ELF DSO
280 The ELF DSO is a pre-linked library that is mapped in by the kernel at
281 the gate page. It is a proper ELF shared object so, with a dynamic
282 loader that recognises the library, you should be able to make calls to
283 the exported functions within it as with any other shared library.
284 AT_SYSINFO points into the kernel DSO at the
285 __kernel_syscall_via_epc() function for historical reasons (it was
286 used before the kernel DSO) and as a convenience.