linux/arch/s390/kernel/ptrace.c

1717 lines
45 KiB
C
Raw Normal View History

License cleanup: add SPDX GPL-2.0 license identifier to files with no license Many source files in the tree are missing licensing information, which makes it harder for compliance tools to determine the correct license. By default all files without license information are under the default license of the kernel, which is GPL version 2. Update the files which contain no license information with the 'GPL-2.0' SPDX license identifier. The SPDX identifier is a legally binding shorthand, which can be used instead of the full boiler plate text. This patch is based on work done by Thomas Gleixner and Kate Stewart and Philippe Ombredanne. How this work was done: Patches were generated and checked against linux-4.14-rc6 for a subset of the use cases: - file had no licensing information it it. - file was a */uapi/* one with no licensing information in it, - file was a */uapi/* one with existing licensing information, Further patches will be generated in subsequent months to fix up cases where non-standard license headers were used, and references to license had to be inferred by heuristics based on keywords. The analysis to determine which SPDX License Identifier to be applied to a file was done in a spreadsheet of side by side results from of the output of two independent scanners (ScanCode & Windriver) producing SPDX tag:value files created by Philippe Ombredanne. Philippe prepared the base worksheet, and did an initial spot review of a few 1000 files. The 4.13 kernel was the starting point of the analysis with 60,537 files assessed. Kate Stewart did a file by file comparison of the scanner results in the spreadsheet to determine which SPDX license identifier(s) to be applied to the file. She confirmed any determination that was not immediately clear with lawyers working with the Linux Foundation. Criteria used to select files for SPDX license identifier tagging was: - Files considered eligible had to be source code files. - Make and config files were included as candidates if they contained >5 lines of source - File already had some variant of a license header in it (even if <5 lines). All documentation files were explicitly excluded. The following heuristics were used to determine which SPDX license identifiers to apply. - when both scanners couldn't find any license traces, file was considered to have no license information in it, and the top level COPYING file license applied. For non */uapi/* files that summary was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 11139 and resulted in the first patch in this series. If that file was a */uapi/* path one, it was "GPL-2.0 WITH Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was: SPDX license identifier # files ---------------------------------------------------|------- GPL-2.0 WITH Linux-syscall-note 930 and resulted in the second patch in this series. - if a file had some form of licensing information in it, and was one of the */uapi/* ones, it was denoted with the Linux-syscall-note if any GPL family license was found in the file or had no licensing in it (per prior point). Results summary: SPDX license identifier # files ---------------------------------------------------|------ GPL-2.0 WITH Linux-syscall-note 270 GPL-2.0+ WITH Linux-syscall-note 169 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21 ((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17 LGPL-2.1+ WITH Linux-syscall-note 15 GPL-1.0+ WITH Linux-syscall-note 14 ((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5 LGPL-2.0+ WITH Linux-syscall-note 4 LGPL-2.1 WITH Linux-syscall-note 3 ((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3 ((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1 and that resulted in the third patch in this series. - when the two scanners agreed on the detected license(s), that became the concluded license(s). - when there was disagreement between the two scanners (one detected a license but the other didn't, or they both detected different licenses) a manual inspection of the file occurred. - In most cases a manual inspection of the information in the file resulted in a clear resolution of the license that should apply (and which scanner probably needed to revisit its heuristics). - When it was not immediately clear, the license identifier was confirmed with lawyers working with the Linux Foundation. - If there was any question as to the appropriate license identifier, the file was flagged for further research and to be revisited later in time. In total, over 70 hours of logged manual review was done on the spreadsheet to determine the SPDX license identifiers to apply to the source files by Kate, Philippe, Thomas and, in some cases, confirmation by lawyers working with the Linux Foundation. Kate also obtained a third independent scan of the 4.13 code base from FOSSology, and compared selected files where the other two scanners disagreed against that SPDX file, to see if there was new insights. The Windriver scanner is based on an older version of FOSSology in part, so they are related. Thomas did random spot checks in about 500 files from the spreadsheets for the uapi headers and agreed with SPDX license identifier in the files he inspected. For the non-uapi files Thomas did random spot checks in about 15000 files. In initial set of patches against 4.14-rc6, 3 files were found to have copy/paste license identifier errors, and have been fixed to reflect the correct identifier. Additionally Philippe spent 10 hours this week doing a detailed manual inspection and review of the 12,461 patched files from the initial patch version early this week with: - a full scancode scan run, collecting the matched texts, detected license ids and scores - reviewing anything where there was a license detected (about 500+ files) to ensure that the applied SPDX license was correct - reviewing anything where there was no detection but the patch license was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied SPDX license was correct This produced a worksheet with 20 files needing minor correction. This worksheet was then exported into 3 different .csv files for the different types of files to be modified. These .csv files were then reviewed by Greg. Thomas wrote a script to parse the csv files and add the proper SPDX tag to the file, in the format that the file expected. This script was further refined by Greg based on the output to detect more types of files automatically and to distinguish between header and source .c files (which need different comment types.) Finally Greg ran the script using the .csv files to generate the patches. Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org> Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-11-01 22:07:57 +08:00
// SPDX-License-Identifier: GPL-2.0
/*
* Ptrace user space interface.
*
* Copyright IBM Corp. 1999, 2010
* Author(s): Denis Joseph Barrow
* Martin Schwidefsky (schwidefsky@de.ibm.com)
*/
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/sched/task_stack.h>
#include <linux/mm.h>
#include <linux/smp.h>
#include <linux/errno.h>
#include <linux/ptrace.h>
#include <linux/user.h>
#include <linux/security.h>
#include <linux/audit.h>
#include <linux/signal.h>
#include <linux/elf.h>
#include <linux/regset.h>
#include <linux/tracehook.h>
#include <linux/seccomp.h>
#include <linux/compat.h>
#include <trace/syscall.h>
#include <asm/page.h>
#include <linux/uaccess.h>
#include <asm/unistd.h>
#include <asm/switch_to.h>
#include <asm/runtime_instr.h>
#include <asm/facility.h>
#include "entry.h"
#ifdef CONFIG_COMPAT
#include "compat_ptrace.h"
#endif
#define CREATE_TRACE_POINTS
#include <trace/events/syscalls.h>
void update_cr_regs(struct task_struct *task)
{
struct pt_regs *regs = task_pt_regs(task);
struct thread_struct *thread = &task->thread;
struct per_regs old, new;
union ctlreg0 cr0_old, cr0_new;
union ctlreg2 cr2_old, cr2_new;
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
int cr0_changed, cr2_changed;
__ctl_store(cr0_old.val, 0, 0);
__ctl_store(cr2_old.val, 2, 2);
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
cr0_new = cr0_old;
cr2_new = cr2_old;
/* Take care of the enable/disable of transactional execution. */
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
if (MACHINE_HAS_TE) {
/* Set or clear transaction execution TXC bit 8. */
cr0_new.tcx = 1;
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
if (task->thread.per_flags & PER_FLAG_NO_TE)
cr0_new.tcx = 0;
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
/* Set or clear transaction execution TDC bits 62 and 63. */
cr2_new.tdc = 0;
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND) {
if (task->thread.per_flags & PER_FLAG_TE_ABORT_RAND_TEND)
cr2_new.tdc = 1;
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
else
cr2_new.tdc = 2;
}
}
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
/* Take care of enable/disable of guarded storage. */
if (MACHINE_HAS_GS) {
cr2_new.gse = 0;
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
if (task->thread.gs_cb)
cr2_new.gse = 1;
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
}
/* Load control register 0/2 iff changed */
cr0_changed = cr0_new.val != cr0_old.val;
cr2_changed = cr2_new.val != cr2_old.val;
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
if (cr0_changed)
__ctl_load(cr0_new.val, 0, 0);
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
if (cr2_changed)
__ctl_load(cr2_new.val, 2, 2);
/* Copy user specified PER registers */
new.control = thread->per_user.control;
new.start = thread->per_user.start;
new.end = thread->per_user.end;
/* merge TIF_SINGLE_STEP into user specified PER registers. */
if (test_tsk_thread_flag(task, TIF_SINGLE_STEP) ||
test_tsk_thread_flag(task, TIF_UPROBE_SINGLESTEP)) {
if (test_tsk_thread_flag(task, TIF_BLOCK_STEP))
new.control |= PER_EVENT_BRANCH;
else
new.control |= PER_EVENT_IFETCH;
new.control |= PER_CONTROL_SUSPENSION;
new.control |= PER_EVENT_TRANSACTION_END;
if (test_tsk_thread_flag(task, TIF_UPROBE_SINGLESTEP))
new.control |= PER_EVENT_IFETCH;
new.start = 0;
new.end = -1UL;
}
/* Take care of the PER enablement bit in the PSW. */
if (!(new.control & PER_EVENT_MASK)) {
regs->psw.mask &= ~PSW_MASK_PER;
return;
}
regs->psw.mask |= PSW_MASK_PER;
__ctl_store(old, 9, 11);
if (memcmp(&new, &old, sizeof(struct per_regs)) != 0)
__ctl_load(new, 9, 11);
}
void user_enable_single_step(struct task_struct *task)
{
clear_tsk_thread_flag(task, TIF_BLOCK_STEP);
set_tsk_thread_flag(task, TIF_SINGLE_STEP);
}
void user_disable_single_step(struct task_struct *task)
{
clear_tsk_thread_flag(task, TIF_BLOCK_STEP);
clear_tsk_thread_flag(task, TIF_SINGLE_STEP);
}
void user_enable_block_step(struct task_struct *task)
{
set_tsk_thread_flag(task, TIF_SINGLE_STEP);
set_tsk_thread_flag(task, TIF_BLOCK_STEP);
}
/*
* Called by kernel/ptrace.c when detaching..
*
* Clear all debugging related fields.
*/
void ptrace_disable(struct task_struct *task)
{
memset(&task->thread.per_user, 0, sizeof(task->thread.per_user));
memset(&task->thread.per_event, 0, sizeof(task->thread.per_event));
clear_tsk_thread_flag(task, TIF_SINGLE_STEP);
clear_pt_regs_flag(task_pt_regs(task), PIF_PER_TRAP);
task->thread.per_flags = 0;
}
#define __ADDR_MASK 7
static inline unsigned long __peek_user_per(struct task_struct *child,
addr_t addr)
{
struct per_struct_kernel *dummy = NULL;
if (addr == (addr_t) &dummy->cr9)
/* Control bits of the active per set. */
return test_thread_flag(TIF_SINGLE_STEP) ?
PER_EVENT_IFETCH : child->thread.per_user.control;
else if (addr == (addr_t) &dummy->cr10)
/* Start address of the active per set. */
return test_thread_flag(TIF_SINGLE_STEP) ?
0 : child->thread.per_user.start;
else if (addr == (addr_t) &dummy->cr11)
/* End address of the active per set. */
return test_thread_flag(TIF_SINGLE_STEP) ?
-1UL : child->thread.per_user.end;
else if (addr == (addr_t) &dummy->bits)
/* Single-step bit. */
return test_thread_flag(TIF_SINGLE_STEP) ?
(1UL << (BITS_PER_LONG - 1)) : 0;
else if (addr == (addr_t) &dummy->starting_addr)
/* Start address of the user specified per set. */
return child->thread.per_user.start;
else if (addr == (addr_t) &dummy->ending_addr)
/* End address of the user specified per set. */
return child->thread.per_user.end;
else if (addr == (addr_t) &dummy->perc_atmid)
/* PER code, ATMID and AI of the last PER trap */
return (unsigned long)
child->thread.per_event.cause << (BITS_PER_LONG - 16);
else if (addr == (addr_t) &dummy->address)
/* Address of the last PER trap */
return child->thread.per_event.address;
else if (addr == (addr_t) &dummy->access_id)
/* Access id of the last PER trap */
return (unsigned long)
child->thread.per_event.paid << (BITS_PER_LONG - 8);
return 0;
}
/*
* Read the word at offset addr from the user area of a process. The
* trouble here is that the information is littered over different
* locations. The process registers are found on the kernel stack,
* the floating point stuff and the trace settings are stored in
* the task structure. In addition the different structures in
* struct user contain pad bytes that should be read as zeroes.
* Lovely...
*/
static unsigned long __peek_user(struct task_struct *child, addr_t addr)
{
struct user *dummy = NULL;
addr_t offset, tmp;
if (addr < (addr_t) &dummy->regs.acrs) {
/*
* psw and gprs are stored on the stack
*/
tmp = *(addr_t *)((addr_t) &task_pt_regs(child)->psw + addr);
if (addr == (addr_t) &dummy->regs.psw.mask) {
/* Return a clean psw mask. */
tmp &= PSW_MASK_USER | PSW_MASK_RI;
tmp |= PSW_USER_BITS;
}
} else if (addr < (addr_t) &dummy->regs.orig_gpr2) {
/*
* access registers are stored in the thread structure
*/
offset = addr - (addr_t) &dummy->regs.acrs;
/*
* Very special case: old & broken 64 bit gdb reading
* from acrs[15]. Result is a 64 bit value. Read the
* 32 bit acrs[15] value and shift it by 32. Sick...
*/
if (addr == (addr_t) &dummy->regs.acrs[15])
tmp = ((unsigned long) child->thread.acrs[15]) << 32;
else
tmp = *(addr_t *)((addr_t) &child->thread.acrs + offset);
} else if (addr == (addr_t) &dummy->regs.orig_gpr2) {
/*
* orig_gpr2 is stored on the kernel stack
*/
tmp = (addr_t) task_pt_regs(child)->orig_gpr2;
} else if (addr < (addr_t) &dummy->regs.fp_regs) {
/*
* prevent reads of padding hole between
* orig_gpr2 and fp_regs on s390.
*/
tmp = 0;
} else if (addr == (addr_t) &dummy->regs.fp_regs.fpc) {
/*
* floating point control reg. is in the thread structure
*/
tmp = child->thread.fpu.fpc;
tmp <<= BITS_PER_LONG - 32;
} else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
/*
* floating point regs. are either in child->thread.fpu
* or the child->thread.fpu.vxrs array
*/
offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
if (MACHINE_HAS_VX)
tmp = *(addr_t *)
((addr_t) child->thread.fpu.vxrs + 2*offset);
else
tmp = *(addr_t *)
((addr_t) child->thread.fpu.fprs + offset);
} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
/*
* Handle access to the per_info structure.
*/
addr -= (addr_t) &dummy->regs.per_info;
tmp = __peek_user_per(child, addr);
} else
tmp = 0;
return tmp;
}
static int
peek_user(struct task_struct *child, addr_t addr, addr_t data)
{
addr_t tmp, mask;
/*
* Stupid gdb peeks/pokes the access registers in 64 bit with
* an alignment of 4. Programmers from hell...
*/
mask = __ADDR_MASK;
if (addr >= (addr_t) &((struct user *) NULL)->regs.acrs &&
addr < (addr_t) &((struct user *) NULL)->regs.orig_gpr2)
mask = 3;
if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK)
return -EIO;
tmp = __peek_user(child, addr);
return put_user(tmp, (addr_t __user *) data);
}
static inline void __poke_user_per(struct task_struct *child,
addr_t addr, addr_t data)
{
struct per_struct_kernel *dummy = NULL;
/*
* There are only three fields in the per_info struct that the
* debugger user can write to.
* 1) cr9: the debugger wants to set a new PER event mask
* 2) starting_addr: the debugger wants to set a new starting
* address to use with the PER event mask.
* 3) ending_addr: the debugger wants to set a new ending
* address to use with the PER event mask.
* The user specified PER event mask and the start and end
* addresses are used only if single stepping is not in effect.
* Writes to any other field in per_info are ignored.
*/
if (addr == (addr_t) &dummy->cr9)
/* PER event mask of the user specified per set. */
child->thread.per_user.control =
data & (PER_EVENT_MASK | PER_CONTROL_MASK);
else if (addr == (addr_t) &dummy->starting_addr)
/* Starting address of the user specified per set. */
child->thread.per_user.start = data;
else if (addr == (addr_t) &dummy->ending_addr)
/* Ending address of the user specified per set. */
child->thread.per_user.end = data;
}
static void fixup_int_code(struct task_struct *child, addr_t data)
{
struct pt_regs *regs = task_pt_regs(child);
int ilc = regs->int_code >> 16;
u16 insn;
if (ilc > 6)
return;
if (ptrace_access_vm(child, regs->psw.addr - (regs->int_code >> 16),
&insn, sizeof(insn), FOLL_FORCE) != sizeof(insn))
return;
/* double check that tracee stopped on svc instruction */
if ((insn >> 8) != 0xa)
return;
regs->int_code = 0x20000 | (data & 0xffff);
}
/*
* Write a word to the user area of a process at location addr. This
* operation does have an additional problem compared to peek_user.
* Stores to the program status word and on the floating point
* control register needs to get checked for validity.
*/
static int __poke_user(struct task_struct *child, addr_t addr, addr_t data)
{
struct user *dummy = NULL;
addr_t offset;
if (addr < (addr_t) &dummy->regs.acrs) {
struct pt_regs *regs = task_pt_regs(child);
/*
* psw and gprs are stored on the stack
*/
if (addr == (addr_t) &dummy->regs.psw.mask) {
unsigned long mask = PSW_MASK_USER;
mask |= is_ri_task(child) ? PSW_MASK_RI : 0;
if ((data ^ PSW_USER_BITS) & ~mask)
/* Invalid psw mask. */
return -EINVAL;
if ((data & PSW_MASK_ASC) == PSW_ASC_HOME)
/* Invalid address-space-control bits */
return -EINVAL;
if ((data & PSW_MASK_EA) && !(data & PSW_MASK_BA))
/* Invalid addressing mode bits */
return -EINVAL;
}
if (test_pt_regs_flag(regs, PIF_SYSCALL) &&
addr == offsetof(struct user, regs.gprs[2]))
fixup_int_code(child, data);
*(addr_t *)((addr_t) &regs->psw + addr) = data;
} else if (addr < (addr_t) (&dummy->regs.orig_gpr2)) {
/*
* access registers are stored in the thread structure
*/
offset = addr - (addr_t) &dummy->regs.acrs;
/*
* Very special case: old & broken 64 bit gdb writing
* to acrs[15] with a 64 bit value. Ignore the lower
* half of the value and write the upper 32 bit to
* acrs[15]. Sick...
*/
if (addr == (addr_t) &dummy->regs.acrs[15])
child->thread.acrs[15] = (unsigned int) (data >> 32);
else
*(addr_t *)((addr_t) &child->thread.acrs + offset) = data;
} else if (addr == (addr_t) &dummy->regs.orig_gpr2) {
/*
* orig_gpr2 is stored on the kernel stack
*/
task_pt_regs(child)->orig_gpr2 = data;
} else if (addr < (addr_t) &dummy->regs.fp_regs) {
/*
* prevent writes of padding hole between
* orig_gpr2 and fp_regs on s390.
*/
return 0;
} else if (addr == (addr_t) &dummy->regs.fp_regs.fpc) {
/*
* floating point control reg. is in the thread structure
*/
if ((unsigned int) data != 0 ||
test_fp_ctl(data >> (BITS_PER_LONG - 32)))
return -EINVAL;
child->thread.fpu.fpc = data >> (BITS_PER_LONG - 32);
} else if (addr < (addr_t) (&dummy->regs.fp_regs + 1)) {
/*
* floating point regs. are either in child->thread.fpu
* or the child->thread.fpu.vxrs array
*/
offset = addr - (addr_t) &dummy->regs.fp_regs.fprs;
if (MACHINE_HAS_VX)
*(addr_t *)((addr_t)
child->thread.fpu.vxrs + 2*offset) = data;
else
*(addr_t *)((addr_t)
child->thread.fpu.fprs + offset) = data;
} else if (addr < (addr_t) (&dummy->regs.per_info + 1)) {
/*
* Handle access to the per_info structure.
*/
addr -= (addr_t) &dummy->regs.per_info;
__poke_user_per(child, addr, data);
}
return 0;
}
static int poke_user(struct task_struct *child, addr_t addr, addr_t data)
{
addr_t mask;
/*
* Stupid gdb peeks/pokes the access registers in 64 bit with
* an alignment of 4. Programmers from hell indeed...
*/
mask = __ADDR_MASK;
if (addr >= (addr_t) &((struct user *) NULL)->regs.acrs &&
addr < (addr_t) &((struct user *) NULL)->regs.orig_gpr2)
mask = 3;
if ((addr & mask) || addr > sizeof(struct user) - __ADDR_MASK)
return -EIO;
return __poke_user(child, addr, data);
}
long arch_ptrace(struct task_struct *child, long request,
unsigned long addr, unsigned long data)
{
ptrace_area parea;
int copied, ret;
switch (request) {
case PTRACE_PEEKUSR:
/* read the word at location addr in the USER area. */
return peek_user(child, addr, data);
case PTRACE_POKEUSR:
/* write the word at location addr in the USER area */
return poke_user(child, addr, data);
case PTRACE_PEEKUSR_AREA:
case PTRACE_POKEUSR_AREA:
if (copy_from_user(&parea, (void __force __user *) addr,
sizeof(parea)))
return -EFAULT;
addr = parea.kernel_addr;
data = parea.process_addr;
copied = 0;
while (copied < parea.len) {
if (request == PTRACE_PEEKUSR_AREA)
ret = peek_user(child, addr, data);
else {
addr_t utmp;
if (get_user(utmp,
(addr_t __force __user *) data))
return -EFAULT;
ret = poke_user(child, addr, utmp);
}
if (ret)
return ret;
addr += sizeof(unsigned long);
data += sizeof(unsigned long);
copied += sizeof(unsigned long);
}
return 0;
case PTRACE_GET_LAST_BREAK:
put_user(child->thread.last_break,
(unsigned long __user *) data);
return 0;
case PTRACE_ENABLE_TE:
if (!MACHINE_HAS_TE)
return -EIO;
child->thread.per_flags &= ~PER_FLAG_NO_TE;
return 0;
case PTRACE_DISABLE_TE:
if (!MACHINE_HAS_TE)
return -EIO;
child->thread.per_flags |= PER_FLAG_NO_TE;
child->thread.per_flags &= ~PER_FLAG_TE_ABORT_RAND;
return 0;
case PTRACE_TE_ABORT_RAND:
if (!MACHINE_HAS_TE || (child->thread.per_flags & PER_FLAG_NO_TE))
return -EIO;
switch (data) {
case 0UL:
child->thread.per_flags &= ~PER_FLAG_TE_ABORT_RAND;
break;
case 1UL:
child->thread.per_flags |= PER_FLAG_TE_ABORT_RAND;
child->thread.per_flags |= PER_FLAG_TE_ABORT_RAND_TEND;
break;
case 2UL:
child->thread.per_flags |= PER_FLAG_TE_ABORT_RAND;
child->thread.per_flags &= ~PER_FLAG_TE_ABORT_RAND_TEND;
break;
default:
return -EINVAL;
}
return 0;
default:
return ptrace_request(child, request, addr, data);
}
}
#ifdef CONFIG_COMPAT
/*
* Now the fun part starts... a 31 bit program running in the
* 31 bit emulation tracing another program. PTRACE_PEEKTEXT,
* PTRACE_PEEKDATA, PTRACE_POKETEXT and PTRACE_POKEDATA are easy
* to handle, the difference to the 64 bit versions of the requests
* is that the access is done in multiples of 4 byte instead of
* 8 bytes (sizeof(unsigned long) on 31/64 bit).
* The ugly part are PTRACE_PEEKUSR, PTRACE_PEEKUSR_AREA,
* PTRACE_POKEUSR and PTRACE_POKEUSR_AREA. If the traced program
* is a 31 bit program too, the content of struct user can be
* emulated. A 31 bit program peeking into the struct user of
* a 64 bit program is a no-no.
*/
/*
* Same as peek_user_per but for a 31 bit program.
*/
static inline __u32 __peek_user_per_compat(struct task_struct *child,
addr_t addr)
{
struct compat_per_struct_kernel *dummy32 = NULL;
if (addr == (addr_t) &dummy32->cr9)
/* Control bits of the active per set. */
return (__u32) test_thread_flag(TIF_SINGLE_STEP) ?
PER_EVENT_IFETCH : child->thread.per_user.control;
else if (addr == (addr_t) &dummy32->cr10)
/* Start address of the active per set. */
return (__u32) test_thread_flag(TIF_SINGLE_STEP) ?
0 : child->thread.per_user.start;
else if (addr == (addr_t) &dummy32->cr11)
/* End address of the active per set. */
return test_thread_flag(TIF_SINGLE_STEP) ?
PSW32_ADDR_INSN : child->thread.per_user.end;
else if (addr == (addr_t) &dummy32->bits)
/* Single-step bit. */
return (__u32) test_thread_flag(TIF_SINGLE_STEP) ?
0x80000000 : 0;
else if (addr == (addr_t) &dummy32->starting_addr)
/* Start address of the user specified per set. */
return (__u32) child->thread.per_user.start;
else if (addr == (addr_t) &dummy32->ending_addr)
/* End address of the user specified per set. */
return (__u32) child->thread.per_user.end;
else if (addr == (addr_t) &dummy32->perc_atmid)
/* PER code, ATMID and AI of the last PER trap */
return (__u32) child->thread.per_event.cause << 16;
else if (addr == (addr_t) &dummy32->address)
/* Address of the last PER trap */
return (__u32) child->thread.per_event.address;
else if (addr == (addr_t) &dummy32->access_id)
/* Access id of the last PER trap */
return (__u32) child->thread.per_event.paid << 24;
return 0;
}
/*
* Same as peek_user but for a 31 bit program.
*/
static u32 __peek_user_compat(struct task_struct *child, addr_t addr)
{
struct compat_user *dummy32 = NULL;
addr_t offset;
__u32 tmp;
if (addr < (addr_t) &dummy32->regs.acrs) {
struct pt_regs *regs = task_pt_regs(child);
/*
* psw and gprs are stored on the stack
*/
if (addr == (addr_t) &dummy32->regs.psw.mask) {
/* Fake a 31 bit psw mask. */
tmp = (__u32)(regs->psw.mask >> 32);
tmp &= PSW32_MASK_USER | PSW32_MASK_RI;
tmp |= PSW32_USER_BITS;
} else if (addr == (addr_t) &dummy32->regs.psw.addr) {
/* Fake a 31 bit psw address. */
tmp = (__u32) regs->psw.addr |
(__u32)(regs->psw.mask & PSW_MASK_BA);
} else {
/* gpr 0-15 */
tmp = *(__u32 *)((addr_t) &regs->psw + addr*2 + 4);
}
} else if (addr < (addr_t) (&dummy32->regs.orig_gpr2)) {
/*
* access registers are stored in the thread structure
*/
offset = addr - (addr_t) &dummy32->regs.acrs;
tmp = *(__u32*)((addr_t) &child->thread.acrs + offset);
} else if (addr == (addr_t) (&dummy32->regs.orig_gpr2)) {
/*
* orig_gpr2 is stored on the kernel stack
*/
tmp = *(__u32*)((addr_t) &task_pt_regs(child)->orig_gpr2 + 4);
} else if (addr < (addr_t) &dummy32->regs.fp_regs) {
/*
* prevent reads of padding hole between
* orig_gpr2 and fp_regs on s390.
*/
tmp = 0;
} else if (addr == (addr_t) &dummy32->regs.fp_regs.fpc) {
/*
* floating point control reg. is in the thread structure
*/
tmp = child->thread.fpu.fpc;
} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
/*
* floating point regs. are either in child->thread.fpu
* or the child->thread.fpu.vxrs array
*/
offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
if (MACHINE_HAS_VX)
tmp = *(__u32 *)
((addr_t) child->thread.fpu.vxrs + 2*offset);
else
tmp = *(__u32 *)
((addr_t) child->thread.fpu.fprs + offset);
} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
/*
* Handle access to the per_info structure.
*/
addr -= (addr_t) &dummy32->regs.per_info;
tmp = __peek_user_per_compat(child, addr);
} else
tmp = 0;
return tmp;
}
static int peek_user_compat(struct task_struct *child,
addr_t addr, addr_t data)
{
__u32 tmp;
if (!is_compat_task() || (addr & 3) || addr > sizeof(struct user) - 3)
return -EIO;
tmp = __peek_user_compat(child, addr);
return put_user(tmp, (__u32 __user *) data);
}
/*
* Same as poke_user_per but for a 31 bit program.
*/
static inline void __poke_user_per_compat(struct task_struct *child,
addr_t addr, __u32 data)
{
struct compat_per_struct_kernel *dummy32 = NULL;
if (addr == (addr_t) &dummy32->cr9)
/* PER event mask of the user specified per set. */
child->thread.per_user.control =
data & (PER_EVENT_MASK | PER_CONTROL_MASK);
else if (addr == (addr_t) &dummy32->starting_addr)
/* Starting address of the user specified per set. */
child->thread.per_user.start = data;
else if (addr == (addr_t) &dummy32->ending_addr)
/* Ending address of the user specified per set. */
child->thread.per_user.end = data;
}
/*
* Same as poke_user but for a 31 bit program.
*/
static int __poke_user_compat(struct task_struct *child,
addr_t addr, addr_t data)
{
struct compat_user *dummy32 = NULL;
__u32 tmp = (__u32) data;
addr_t offset;
if (addr < (addr_t) &dummy32->regs.acrs) {
struct pt_regs *regs = task_pt_regs(child);
/*
* psw, gprs, acrs and orig_gpr2 are stored on the stack
*/
if (addr == (addr_t) &dummy32->regs.psw.mask) {
__u32 mask = PSW32_MASK_USER;
mask |= is_ri_task(child) ? PSW32_MASK_RI : 0;
/* Build a 64 bit psw mask from 31 bit mask. */
if ((tmp ^ PSW32_USER_BITS) & ~mask)
/* Invalid psw mask. */
return -EINVAL;
if ((data & PSW32_MASK_ASC) == PSW32_ASC_HOME)
/* Invalid address-space-control bits */
return -EINVAL;
regs->psw.mask = (regs->psw.mask & ~PSW_MASK_USER) |
(regs->psw.mask & PSW_MASK_BA) |
(__u64)(tmp & mask) << 32;
} else if (addr == (addr_t) &dummy32->regs.psw.addr) {
/* Build a 64 bit psw address from 31 bit address. */
regs->psw.addr = (__u64) tmp & PSW32_ADDR_INSN;
/* Transfer 31 bit amode bit to psw mask. */
regs->psw.mask = (regs->psw.mask & ~PSW_MASK_BA) |
(__u64)(tmp & PSW32_ADDR_AMODE);
} else {
if (test_pt_regs_flag(regs, PIF_SYSCALL) &&
addr == offsetof(struct compat_user, regs.gprs[2]))
fixup_int_code(child, data);
/* gpr 0-15 */
*(__u32*)((addr_t) &regs->psw + addr*2 + 4) = tmp;
}
} else if (addr < (addr_t) (&dummy32->regs.orig_gpr2)) {
/*
* access registers are stored in the thread structure
*/
offset = addr - (addr_t) &dummy32->regs.acrs;
*(__u32*)((addr_t) &child->thread.acrs + offset) = tmp;
} else if (addr == (addr_t) (&dummy32->regs.orig_gpr2)) {
/*
* orig_gpr2 is stored on the kernel stack
*/
*(__u32*)((addr_t) &task_pt_regs(child)->orig_gpr2 + 4) = tmp;
} else if (addr < (addr_t) &dummy32->regs.fp_regs) {
/*
* prevent writess of padding hole between
* orig_gpr2 and fp_regs on s390.
*/
return 0;
} else if (addr == (addr_t) &dummy32->regs.fp_regs.fpc) {
/*
* floating point control reg. is in the thread structure
*/
if (test_fp_ctl(tmp))
return -EINVAL;
child->thread.fpu.fpc = data;
} else if (addr < (addr_t) (&dummy32->regs.fp_regs + 1)) {
/*
* floating point regs. are either in child->thread.fpu
* or the child->thread.fpu.vxrs array
*/
offset = addr - (addr_t) &dummy32->regs.fp_regs.fprs;
if (MACHINE_HAS_VX)
*(__u32 *)((addr_t)
child->thread.fpu.vxrs + 2*offset) = tmp;
else
*(__u32 *)((addr_t)
child->thread.fpu.fprs + offset) = tmp;
} else if (addr < (addr_t) (&dummy32->regs.per_info + 1)) {
/*
* Handle access to the per_info structure.
*/
addr -= (addr_t) &dummy32->regs.per_info;
__poke_user_per_compat(child, addr, data);
}
return 0;
}
static int poke_user_compat(struct task_struct *child,
addr_t addr, addr_t data)
{
if (!is_compat_task() || (addr & 3) ||
addr > sizeof(struct compat_user) - 3)
return -EIO;
return __poke_user_compat(child, addr, data);
}
long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
compat_ulong_t caddr, compat_ulong_t cdata)
{
unsigned long addr = caddr;
unsigned long data = cdata;
compat_ptrace_area parea;
int copied, ret;
switch (request) {
case PTRACE_PEEKUSR:
/* read the word at location addr in the USER area. */
return peek_user_compat(child, addr, data);
case PTRACE_POKEUSR:
/* write the word at location addr in the USER area */
return poke_user_compat(child, addr, data);
case PTRACE_PEEKUSR_AREA:
case PTRACE_POKEUSR_AREA:
if (copy_from_user(&parea, (void __force __user *) addr,
sizeof(parea)))
return -EFAULT;
addr = parea.kernel_addr;
data = parea.process_addr;
copied = 0;
while (copied < parea.len) {
if (request == PTRACE_PEEKUSR_AREA)
ret = peek_user_compat(child, addr, data);
else {
__u32 utmp;
if (get_user(utmp,
(__u32 __force __user *) data))
return -EFAULT;
ret = poke_user_compat(child, addr, utmp);
}
if (ret)
return ret;
addr += sizeof(unsigned int);
data += sizeof(unsigned int);
copied += sizeof(unsigned int);
}
return 0;
case PTRACE_GET_LAST_BREAK:
put_user(child->thread.last_break,
(unsigned int __user *) data);
return 0;
}
return compat_ptrace_request(child, request, addr, data);
}
#endif
asmlinkage long do_syscall_trace_enter(struct pt_regs *regs)
{
unsigned long mask = -1UL;
long ret = -1;
if (is_compat_task())
mask = 0xffffffff;
/*
* The sysc_tracesys code in entry.S stored the system
* call number to gprs[2].
*/
if (test_thread_flag(TIF_SYSCALL_TRACE) &&
tracehook_report_syscall_entry(regs)) {
/*
* Tracing decided this syscall should not happen. Skip
* the system call and the system call restart handling.
*/
goto skip;
}
#ifdef CONFIG_SECCOMP
/* Do the secure computing check after ptrace. */
if (unlikely(test_thread_flag(TIF_SECCOMP))) {
struct seccomp_data sd;
if (is_compat_task()) {
sd.instruction_pointer = regs->psw.addr & 0x7fffffff;
sd.arch = AUDIT_ARCH_S390;
} else {
sd.instruction_pointer = regs->psw.addr;
sd.arch = AUDIT_ARCH_S390X;
}
sd.nr = regs->int_code & 0xffff;
sd.args[0] = regs->orig_gpr2 & mask;
sd.args[1] = regs->gprs[3] & mask;
sd.args[2] = regs->gprs[4] & mask;
sd.args[3] = regs->gprs[5] & mask;
sd.args[4] = regs->gprs[6] & mask;
sd.args[5] = regs->gprs[7] & mask;
if (__secure_computing(&sd) == -1)
goto skip;
}
#endif /* CONFIG_SECCOMP */
if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
trace_sys_enter(regs, regs->int_code & 0xffff);
audit_syscall_entry(regs->int_code & 0xffff, regs->orig_gpr2 & mask,
regs->gprs[3] &mask, regs->gprs[4] &mask,
regs->gprs[5] &mask);
if ((signed long)regs->gprs[2] >= NR_syscalls) {
regs->gprs[2] = -ENOSYS;
ret = -ENOSYS;
}
return regs->gprs[2];
skip:
clear_pt_regs_flag(regs, PIF_SYSCALL);
return ret;
}
asmlinkage void do_syscall_trace_exit(struct pt_regs *regs)
{
Audit: push audit success and retcode into arch ptrace.h The audit system previously expected arches calling to audit_syscall_exit to supply as arguments if the syscall was a success and what the return code was. Audit also provides a helper AUDITSC_RESULT which was supposed to simplify things by converting from negative retcodes to an audit internal magic value stating success or failure. This helper was wrong and could indicate that a valid pointer returned to userspace was a failed syscall. The fix is to fix the layering foolishness. We now pass audit_syscall_exit a struct pt_reg and it in turns calls back into arch code to collect the return value and to determine if the syscall was a success or failure. We also define a generic is_syscall_success() macro which determines success/failure based on if the value is < -MAX_ERRNO. This works for arches like x86 which do not use a separate mechanism to indicate syscall failure. We make both the is_syscall_success() and regs_return_value() static inlines instead of macros. The reason is because the audit function must take a void* for the regs. (uml calls theirs struct uml_pt_regs instead of just struct pt_regs so audit_syscall_exit can't take a struct pt_regs). Since the audit function takes a void* we need to use static inlines to cast it back to the arch correct structure to dereference it. The other major change is that on some arches, like ia64, MIPS and ppc, we change regs_return_value() to give us the negative value on syscall failure. THE only other user of this macro, kretprobe_example.c, won't notice and it makes the value signed consistently for the audit functions across all archs. In arch/sh/kernel/ptrace_64.c I see that we were using regs[9] in the old audit code as the return value. But the ptrace_64.h code defined the macro regs_return_value() as regs[3]. I have no idea which one is correct, but this patch now uses the regs_return_value() function, so it now uses regs[3]. For powerpc we previously used regs->result but now use the regs_return_value() function which uses regs->gprs[3]. regs->gprs[3] is always positive so the regs_return_value(), much like ia64 makes it negative before calling the audit code when appropriate. Signed-off-by: Eric Paris <eparis@redhat.com> Acked-by: H. Peter Anvin <hpa@zytor.com> [for x86 portion] Acked-by: Tony Luck <tony.luck@intel.com> [for ia64] Acked-by: Richard Weinberger <richard@nod.at> [for uml] Acked-by: David S. Miller <davem@davemloft.net> [for sparc] Acked-by: Ralf Baechle <ralf@linux-mips.org> [for mips] Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> [for ppc]
2012-01-04 03:23:06 +08:00
audit_syscall_exit(regs);
if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT)))
trace_sys_exit(regs, regs->gprs[2]);
if (test_thread_flag(TIF_SYSCALL_TRACE))
tracehook_report_syscall_exit(regs, 0);
}
/*
* user_regset definitions.
*/
static int s390_regs_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
unsigned pos;
if (target == current)
save_access_regs(target->thread.acrs);
for (pos = 0; pos < sizeof(s390_regs); pos += sizeof(long))
membuf_store(&to, __peek_user(target, pos));
return 0;
}
static int s390_regs_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
int rc = 0;
if (target == current)
save_access_regs(target->thread.acrs);
if (kbuf) {
const unsigned long *k = kbuf;
while (count > 0 && !rc) {
rc = __poke_user(target, pos, *k++);
count -= sizeof(*k);
pos += sizeof(*k);
}
} else {
const unsigned long __user *u = ubuf;
while (count > 0 && !rc) {
unsigned long word;
rc = __get_user(word, u++);
if (rc)
break;
rc = __poke_user(target, pos, word);
count -= sizeof(*u);
pos += sizeof(*u);
}
}
if (rc == 0 && target == current)
restore_access_regs(target->thread.acrs);
return rc;
}
static int s390_fpregs_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
_s390_fp_regs fp_regs;
if (target == current)
save_fpu_regs();
fp_regs.fpc = target->thread.fpu.fpc;
fpregs_store(&fp_regs, &target->thread.fpu);
return membuf_write(&to, &fp_regs, sizeof(fp_regs));
}
static int s390_fpregs_set(struct task_struct *target,
const struct user_regset *regset, unsigned int pos,
unsigned int count, const void *kbuf,
const void __user *ubuf)
{
int rc = 0;
freg_t fprs[__NUM_FPRS];
if (target == current)
save_fpu_regs();
if (MACHINE_HAS_VX)
convert_vx_to_fp(fprs, target->thread.fpu.vxrs);
else
memcpy(&fprs, target->thread.fpu.fprs, sizeof(fprs));
/* If setting FPC, must validate it first. */
if (count > 0 && pos < offsetof(s390_fp_regs, fprs)) {
u32 ufpc[2] = { target->thread.fpu.fpc, 0 };
rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, &ufpc,
0, offsetof(s390_fp_regs, fprs));
if (rc)
return rc;
if (ufpc[1] != 0 || test_fp_ctl(ufpc[0]))
return -EINVAL;
target->thread.fpu.fpc = ufpc[0];
}
if (rc == 0 && count > 0)
rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
fprs, offsetof(s390_fp_regs, fprs), -1);
if (rc)
return rc;
if (MACHINE_HAS_VX)
convert_fp_to_vx(target->thread.fpu.vxrs, fprs);
else
memcpy(target->thread.fpu.fprs, &fprs, sizeof(fprs));
return rc;
}
static int s390_last_break_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
return membuf_store(&to, target->thread.last_break);
}
static int s390_last_break_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
return 0;
}
static int s390_tdb_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
struct pt_regs *regs = task_pt_regs(target);
if (!(regs->int_code & 0x200))
return -ENODATA;
return membuf_write(&to, target->thread.trap_tdb, 256);
}
static int s390_tdb_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
return 0;
}
static int s390_vxrs_low_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
__u64 vxrs[__NUM_VXRS_LOW];
int i;
if (!MACHINE_HAS_VX)
return -ENODEV;
if (target == current)
save_fpu_regs();
for (i = 0; i < __NUM_VXRS_LOW; i++)
vxrs[i] = *((__u64 *)(target->thread.fpu.vxrs + i) + 1);
return membuf_write(&to, vxrs, sizeof(vxrs));
}
static int s390_vxrs_low_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
__u64 vxrs[__NUM_VXRS_LOW];
int i, rc;
if (!MACHINE_HAS_VX)
return -ENODEV;
if (target == current)
save_fpu_regs();
for (i = 0; i < __NUM_VXRS_LOW; i++)
vxrs[i] = *((__u64 *)(target->thread.fpu.vxrs + i) + 1);
rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf, vxrs, 0, -1);
s390/kernel: lazy restore fpu registers Improve the save and restore behavior of FPU register contents to use the vector extension within the kernel. The kernel does not use floating-point or vector registers and, therefore, saving and restoring the FPU register contents are performed for handling signals or switching processes only. To prepare for using vector instructions and vector registers within the kernel, enhance the save behavior and implement a lazy restore at return to user space from a system call or interrupt. To implement the lazy restore, the save_fpu_regs() sets a CPU information flag, CIF_FPU, to indicate that the FPU registers must be restored. Saving and setting CIF_FPU is performed in an atomic fashion to be interrupt-safe. When the kernel wants to use the vector extension or wants to change the FPU register state for a task during signal handling, the save_fpu_regs() must be called first. The CIF_FPU flag is also set at process switch. At return to user space, the FPU state is restored. In particular, the FPU state includes the floating-point or vector register contents, as well as, vector-enablement and floating-point control. The FPU state restore and clearing CIF_FPU is also performed in an atomic fashion. For KVM, the restore of the FPU register state is performed when restoring the general-purpose guest registers before the SIE instructions is started. Because the path towards the SIE instruction is interruptible, the CIF_FPU flag must be checked again right before going into SIE. If set, the guest registers must be reloaded again by re-entering the outer SIE loop. This is the same behavior as if the SIE critical section is interrupted. Signed-off-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com> Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2015-06-10 18:53:42 +08:00
if (rc == 0)
for (i = 0; i < __NUM_VXRS_LOW; i++)
*((__u64 *)(target->thread.fpu.vxrs + i) + 1) = vxrs[i];
return rc;
}
static int s390_vxrs_high_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
if (!MACHINE_HAS_VX)
return -ENODEV;
if (target == current)
save_fpu_regs();
return membuf_write(&to, target->thread.fpu.vxrs + __NUM_VXRS_LOW,
__NUM_VXRS_HIGH * sizeof(__vector128));
}
static int s390_vxrs_high_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
int rc;
if (!MACHINE_HAS_VX)
return -ENODEV;
if (target == current)
save_fpu_regs();
rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
target->thread.fpu.vxrs + __NUM_VXRS_LOW, 0, -1);
return rc;
}
[S390] signal race with restarting system calls For a ERESTARTNOHAND/ERESTARTSYS/ERESTARTNOINTR restarting system call do_signal will prepare the restart of the system call with a rewind of the PSW before calling get_signal_to_deliver (where the debugger might take control). For A ERESTART_RESTARTBLOCK restarting system call do_signal will set -EINTR as return code. There are two issues with this approach: 1) strace never sees ERESTARTNOHAND, ERESTARTSYS, ERESTARTNOINTR or ERESTART_RESTARTBLOCK as the rewinding already took place or the return code has been changed to -EINTR 2) if get_signal_to_deliver does not return with a signal to deliver the restart via the repeat of the svc instruction is left in place. This opens a race if another signal is made pending before the system call instruction can be reexecuted. The original system call will be restarted even if the second signal would have ended the system call with -EINTR. These two issues can be solved by dropping the early rewind of the system call before get_signal_to_deliver has been called and by using the TIF_RESTART_SVC magic to do the restart if no signal has to be delivered. The only situation where the system call restart via the repeat of the svc instruction is appropriate is when a SA_RESTART signal is delivered to user space. Unfortunately this breaks inferior calls by the debugger again. The system call number and the length of the system call instruction is lost over the inferior call and user space will see ERESTARTNOHAND/ ERESTARTSYS/ERESTARTNOINTR/ERESTART_RESTARTBLOCK. To correct this a new ptrace interface is added to save/restore the system call number and system call instruction length. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2011-10-30 22:16:47 +08:00
static int s390_system_call_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
[S390] signal race with restarting system calls For a ERESTARTNOHAND/ERESTARTSYS/ERESTARTNOINTR restarting system call do_signal will prepare the restart of the system call with a rewind of the PSW before calling get_signal_to_deliver (where the debugger might take control). For A ERESTART_RESTARTBLOCK restarting system call do_signal will set -EINTR as return code. There are two issues with this approach: 1) strace never sees ERESTARTNOHAND, ERESTARTSYS, ERESTARTNOINTR or ERESTART_RESTARTBLOCK as the rewinding already took place or the return code has been changed to -EINTR 2) if get_signal_to_deliver does not return with a signal to deliver the restart via the repeat of the svc instruction is left in place. This opens a race if another signal is made pending before the system call instruction can be reexecuted. The original system call will be restarted even if the second signal would have ended the system call with -EINTR. These two issues can be solved by dropping the early rewind of the system call before get_signal_to_deliver has been called and by using the TIF_RESTART_SVC magic to do the restart if no signal has to be delivered. The only situation where the system call restart via the repeat of the svc instruction is appropriate is when a SA_RESTART signal is delivered to user space. Unfortunately this breaks inferior calls by the debugger again. The system call number and the length of the system call instruction is lost over the inferior call and user space will see ERESTARTNOHAND/ ERESTARTSYS/ERESTARTNOINTR/ERESTART_RESTARTBLOCK. To correct this a new ptrace interface is added to save/restore the system call number and system call instruction length. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2011-10-30 22:16:47 +08:00
{
return membuf_store(&to, target->thread.system_call);
[S390] signal race with restarting system calls For a ERESTARTNOHAND/ERESTARTSYS/ERESTARTNOINTR restarting system call do_signal will prepare the restart of the system call with a rewind of the PSW before calling get_signal_to_deliver (where the debugger might take control). For A ERESTART_RESTARTBLOCK restarting system call do_signal will set -EINTR as return code. There are two issues with this approach: 1) strace never sees ERESTARTNOHAND, ERESTARTSYS, ERESTARTNOINTR or ERESTART_RESTARTBLOCK as the rewinding already took place or the return code has been changed to -EINTR 2) if get_signal_to_deliver does not return with a signal to deliver the restart via the repeat of the svc instruction is left in place. This opens a race if another signal is made pending before the system call instruction can be reexecuted. The original system call will be restarted even if the second signal would have ended the system call with -EINTR. These two issues can be solved by dropping the early rewind of the system call before get_signal_to_deliver has been called and by using the TIF_RESTART_SVC magic to do the restart if no signal has to be delivered. The only situation where the system call restart via the repeat of the svc instruction is appropriate is when a SA_RESTART signal is delivered to user space. Unfortunately this breaks inferior calls by the debugger again. The system call number and the length of the system call instruction is lost over the inferior call and user space will see ERESTARTNOHAND/ ERESTARTSYS/ERESTARTNOINTR/ERESTART_RESTARTBLOCK. To correct this a new ptrace interface is added to save/restore the system call number and system call instruction length. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2011-10-30 22:16:47 +08:00
}
static int s390_system_call_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
unsigned int *data = &target->thread.system_call;
[S390] signal race with restarting system calls For a ERESTARTNOHAND/ERESTARTSYS/ERESTARTNOINTR restarting system call do_signal will prepare the restart of the system call with a rewind of the PSW before calling get_signal_to_deliver (where the debugger might take control). For A ERESTART_RESTARTBLOCK restarting system call do_signal will set -EINTR as return code. There are two issues with this approach: 1) strace never sees ERESTARTNOHAND, ERESTARTSYS, ERESTARTNOINTR or ERESTART_RESTARTBLOCK as the rewinding already took place or the return code has been changed to -EINTR 2) if get_signal_to_deliver does not return with a signal to deliver the restart via the repeat of the svc instruction is left in place. This opens a race if another signal is made pending before the system call instruction can be reexecuted. The original system call will be restarted even if the second signal would have ended the system call with -EINTR. These two issues can be solved by dropping the early rewind of the system call before get_signal_to_deliver has been called and by using the TIF_RESTART_SVC magic to do the restart if no signal has to be delivered. The only situation where the system call restart via the repeat of the svc instruction is appropriate is when a SA_RESTART signal is delivered to user space. Unfortunately this breaks inferior calls by the debugger again. The system call number and the length of the system call instruction is lost over the inferior call and user space will see ERESTARTNOHAND/ ERESTARTSYS/ERESTARTNOINTR/ERESTART_RESTARTBLOCK. To correct this a new ptrace interface is added to save/restore the system call number and system call instruction length. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2011-10-30 22:16:47 +08:00
return user_regset_copyin(&pos, &count, &kbuf, &ubuf,
data, 0, sizeof(unsigned int));
}
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
static int s390_gs_cb_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
{
struct gs_cb *data = target->thread.gs_cb;
if (!MACHINE_HAS_GS)
return -ENODEV;
if (!data)
return -ENODATA;
if (target == current)
save_gs_cb(data);
return membuf_write(&to, data, sizeof(struct gs_cb));
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
}
static int s390_gs_cb_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
struct gs_cb gs_cb = { }, *data = NULL;
int rc;
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
if (!MACHINE_HAS_GS)
return -ENODEV;
if (!target->thread.gs_cb) {
data = kzalloc(sizeof(*data), GFP_KERNEL);
if (!data)
return -ENOMEM;
}
if (!target->thread.gs_cb)
gs_cb.gsd = 25;
else if (target == current)
save_gs_cb(&gs_cb);
else
gs_cb = *target->thread.gs_cb;
rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
&gs_cb, 0, sizeof(gs_cb));
if (rc) {
kfree(data);
return -EFAULT;
}
preempt_disable();
if (!target->thread.gs_cb)
target->thread.gs_cb = data;
*target->thread.gs_cb = gs_cb;
if (target == current) {
__ctl_set_bit(2, 4);
restore_gs_cb(target->thread.gs_cb);
}
preempt_enable();
return rc;
}
static int s390_gs_bc_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
struct gs_cb *data = target->thread.gs_bc_cb;
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
if (!MACHINE_HAS_GS)
return -ENODEV;
if (!data)
return -ENODATA;
return membuf_write(&to, data, sizeof(struct gs_cb));
}
static int s390_gs_bc_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
struct gs_cb *data = target->thread.gs_bc_cb;
if (!MACHINE_HAS_GS)
return -ENODEV;
if (!data) {
data = kzalloc(sizeof(*data), GFP_KERNEL);
if (!data)
return -ENOMEM;
target->thread.gs_bc_cb = data;
}
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
return user_regset_copyin(&pos, &count, &kbuf, &ubuf,
data, 0, sizeof(struct gs_cb));
}
static bool is_ri_cb_valid(struct runtime_instr_cb *cb)
{
return (cb->rca & 0x1f) == 0 &&
(cb->roa & 0xfff) == 0 &&
(cb->rla & 0xfff) == 0xfff &&
cb->s == 1 &&
cb->k == 1 &&
cb->h == 0 &&
cb->reserved1 == 0 &&
cb->ps == 1 &&
cb->qs == 0 &&
cb->pc == 1 &&
cb->qc == 0 &&
cb->reserved2 == 0 &&
cb->reserved3 == 0 &&
cb->reserved4 == 0 &&
cb->reserved5 == 0 &&
cb->reserved6 == 0 &&
cb->reserved7 == 0 &&
cb->reserved8 == 0 &&
cb->rla >= cb->roa &&
cb->rca >= cb->roa &&
cb->rca <= cb->rla+1 &&
cb->m < 3;
}
static int s390_runtime_instr_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
struct runtime_instr_cb *data = target->thread.ri_cb;
if (!test_facility(64))
return -ENODEV;
if (!data)
return -ENODATA;
return membuf_write(&to, data, sizeof(struct runtime_instr_cb));
}
static int s390_runtime_instr_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
struct runtime_instr_cb ri_cb = { }, *data = NULL;
int rc;
if (!test_facility(64))
return -ENODEV;
if (!target->thread.ri_cb) {
data = kzalloc(sizeof(*data), GFP_KERNEL);
if (!data)
return -ENOMEM;
}
if (target->thread.ri_cb) {
if (target == current)
store_runtime_instr_cb(&ri_cb);
else
ri_cb = *target->thread.ri_cb;
}
rc = user_regset_copyin(&pos, &count, &kbuf, &ubuf,
&ri_cb, 0, sizeof(struct runtime_instr_cb));
if (rc) {
kfree(data);
return -EFAULT;
}
if (!is_ri_cb_valid(&ri_cb)) {
kfree(data);
return -EINVAL;
}
/*
* Override access key in any case, since user space should
* not be able to set it, nor should it care about it.
*/
ri_cb.key = PAGE_DEFAULT_KEY >> 4;
preempt_disable();
if (!target->thread.ri_cb)
target->thread.ri_cb = data;
*target->thread.ri_cb = ri_cb;
if (target == current)
load_runtime_instr_cb(target->thread.ri_cb);
preempt_enable();
return 0;
}
static const struct user_regset s390_regsets[] = {
{
.core_note_type = NT_PRSTATUS,
.n = sizeof(s390_regs) / sizeof(long),
.size = sizeof(long),
.align = sizeof(long),
.regset_get = s390_regs_get,
.set = s390_regs_set,
},
{
.core_note_type = NT_PRFPREG,
.n = sizeof(s390_fp_regs) / sizeof(long),
.size = sizeof(long),
.align = sizeof(long),
.regset_get = s390_fpregs_get,
.set = s390_fpregs_set,
},
{
.core_note_type = NT_S390_SYSTEM_CALL,
.n = 1,
.size = sizeof(unsigned int),
.align = sizeof(unsigned int),
.regset_get = s390_system_call_get,
.set = s390_system_call_set,
},
{
.core_note_type = NT_S390_LAST_BREAK,
.n = 1,
.size = sizeof(long),
.align = sizeof(long),
.regset_get = s390_last_break_get,
.set = s390_last_break_set,
},
{
.core_note_type = NT_S390_TDB,
.n = 1,
.size = 256,
.align = 1,
.regset_get = s390_tdb_get,
.set = s390_tdb_set,
},
{
.core_note_type = NT_S390_VXRS_LOW,
.n = __NUM_VXRS_LOW,
.size = sizeof(__u64),
.align = sizeof(__u64),
.regset_get = s390_vxrs_low_get,
.set = s390_vxrs_low_set,
[S390] signal race with restarting system calls For a ERESTARTNOHAND/ERESTARTSYS/ERESTARTNOINTR restarting system call do_signal will prepare the restart of the system call with a rewind of the PSW before calling get_signal_to_deliver (where the debugger might take control). For A ERESTART_RESTARTBLOCK restarting system call do_signal will set -EINTR as return code. There are two issues with this approach: 1) strace never sees ERESTARTNOHAND, ERESTARTSYS, ERESTARTNOINTR or ERESTART_RESTARTBLOCK as the rewinding already took place or the return code has been changed to -EINTR 2) if get_signal_to_deliver does not return with a signal to deliver the restart via the repeat of the svc instruction is left in place. This opens a race if another signal is made pending before the system call instruction can be reexecuted. The original system call will be restarted even if the second signal would have ended the system call with -EINTR. These two issues can be solved by dropping the early rewind of the system call before get_signal_to_deliver has been called and by using the TIF_RESTART_SVC magic to do the restart if no signal has to be delivered. The only situation where the system call restart via the repeat of the svc instruction is appropriate is when a SA_RESTART signal is delivered to user space. Unfortunately this breaks inferior calls by the debugger again. The system call number and the length of the system call instruction is lost over the inferior call and user space will see ERESTARTNOHAND/ ERESTARTSYS/ERESTARTNOINTR/ERESTART_RESTARTBLOCK. To correct this a new ptrace interface is added to save/restore the system call number and system call instruction length. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2011-10-30 22:16:47 +08:00
},
{
.core_note_type = NT_S390_VXRS_HIGH,
.n = __NUM_VXRS_HIGH,
.size = sizeof(__vector128),
.align = sizeof(__vector128),
.regset_get = s390_vxrs_high_get,
.set = s390_vxrs_high_set,
[S390] signal race with restarting system calls For a ERESTARTNOHAND/ERESTARTSYS/ERESTARTNOINTR restarting system call do_signal will prepare the restart of the system call with a rewind of the PSW before calling get_signal_to_deliver (where the debugger might take control). For A ERESTART_RESTARTBLOCK restarting system call do_signal will set -EINTR as return code. There are two issues with this approach: 1) strace never sees ERESTARTNOHAND, ERESTARTSYS, ERESTARTNOINTR or ERESTART_RESTARTBLOCK as the rewinding already took place or the return code has been changed to -EINTR 2) if get_signal_to_deliver does not return with a signal to deliver the restart via the repeat of the svc instruction is left in place. This opens a race if another signal is made pending before the system call instruction can be reexecuted. The original system call will be restarted even if the second signal would have ended the system call with -EINTR. These two issues can be solved by dropping the early rewind of the system call before get_signal_to_deliver has been called and by using the TIF_RESTART_SVC magic to do the restart if no signal has to be delivered. The only situation where the system call restart via the repeat of the svc instruction is appropriate is when a SA_RESTART signal is delivered to user space. Unfortunately this breaks inferior calls by the debugger again. The system call number and the length of the system call instruction is lost over the inferior call and user space will see ERESTARTNOHAND/ ERESTARTSYS/ERESTARTNOINTR/ERESTART_RESTARTBLOCK. To correct this a new ptrace interface is added to save/restore the system call number and system call instruction length. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2011-10-30 22:16:47 +08:00
},
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
{
.core_note_type = NT_S390_GS_CB,
.n = sizeof(struct gs_cb) / sizeof(__u64),
.size = sizeof(__u64),
.align = sizeof(__u64),
.regset_get = s390_gs_cb_get,
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
.set = s390_gs_cb_set,
},
{
.core_note_type = NT_S390_GS_BC,
.n = sizeof(struct gs_cb) / sizeof(__u64),
.size = sizeof(__u64),
.align = sizeof(__u64),
.regset_get = s390_gs_bc_get,
.set = s390_gs_bc_set,
},
{
.core_note_type = NT_S390_RI_CB,
.n = sizeof(struct runtime_instr_cb) / sizeof(__u64),
.size = sizeof(__u64),
.align = sizeof(__u64),
.regset_get = s390_runtime_instr_get,
.set = s390_runtime_instr_set,
},
};
static const struct user_regset_view user_s390_view = {
.name = "s390x",
.e_machine = EM_S390,
.regsets = s390_regsets,
.n = ARRAY_SIZE(s390_regsets)
};
#ifdef CONFIG_COMPAT
static int s390_compat_regs_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
unsigned n;
if (target == current)
save_access_regs(target->thread.acrs);
for (n = 0; n < sizeof(s390_compat_regs); n += sizeof(compat_ulong_t))
membuf_store(&to, __peek_user_compat(target, n));
return 0;
}
static int s390_compat_regs_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
int rc = 0;
if (target == current)
save_access_regs(target->thread.acrs);
if (kbuf) {
const compat_ulong_t *k = kbuf;
while (count > 0 && !rc) {
rc = __poke_user_compat(target, pos, *k++);
count -= sizeof(*k);
pos += sizeof(*k);
}
} else {
const compat_ulong_t __user *u = ubuf;
while (count > 0 && !rc) {
compat_ulong_t word;
rc = __get_user(word, u++);
if (rc)
break;
rc = __poke_user_compat(target, pos, word);
count -= sizeof(*u);
pos += sizeof(*u);
}
}
if (rc == 0 && target == current)
restore_access_regs(target->thread.acrs);
return rc;
}
static int s390_compat_regs_high_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
compat_ulong_t *gprs_high;
int i;
gprs_high = (compat_ulong_t *)task_pt_regs(target)->gprs;
for (i = 0; i < NUM_GPRS; i++, gprs_high += 2)
membuf_store(&to, *gprs_high);
return 0;
}
static int s390_compat_regs_high_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
compat_ulong_t *gprs_high;
int rc = 0;
gprs_high = (compat_ulong_t *)
&task_pt_regs(target)->gprs[pos / sizeof(compat_ulong_t)];
if (kbuf) {
const compat_ulong_t *k = kbuf;
while (count > 0) {
*gprs_high = *k++;
*gprs_high += 2;
count -= sizeof(*k);
}
} else {
const compat_ulong_t __user *u = ubuf;
while (count > 0 && !rc) {
unsigned long word;
rc = __get_user(word, u++);
if (rc)
break;
*gprs_high = word;
*gprs_high += 2;
count -= sizeof(*u);
}
}
return rc;
}
static int s390_compat_last_break_get(struct task_struct *target,
const struct user_regset *regset,
struct membuf to)
{
compat_ulong_t last_break = target->thread.last_break;
return membuf_store(&to, (unsigned long)last_break);
}
static int s390_compat_last_break_set(struct task_struct *target,
const struct user_regset *regset,
unsigned int pos, unsigned int count,
const void *kbuf, const void __user *ubuf)
{
return 0;
}
static const struct user_regset s390_compat_regsets[] = {
{
.core_note_type = NT_PRSTATUS,
.n = sizeof(s390_compat_regs) / sizeof(compat_long_t),
.size = sizeof(compat_long_t),
.align = sizeof(compat_long_t),
.regset_get = s390_compat_regs_get,
.set = s390_compat_regs_set,
},
{
.core_note_type = NT_PRFPREG,
.n = sizeof(s390_fp_regs) / sizeof(compat_long_t),
.size = sizeof(compat_long_t),
.align = sizeof(compat_long_t),
.regset_get = s390_fpregs_get,
.set = s390_fpregs_set,
},
{
.core_note_type = NT_S390_SYSTEM_CALL,
.n = 1,
.size = sizeof(compat_uint_t),
.align = sizeof(compat_uint_t),
.regset_get = s390_system_call_get,
.set = s390_system_call_set,
},
{
.core_note_type = NT_S390_LAST_BREAK,
.n = 1,
.size = sizeof(long),
.align = sizeof(long),
.regset_get = s390_compat_last_break_get,
.set = s390_compat_last_break_set,
},
{
.core_note_type = NT_S390_TDB,
.n = 1,
.size = 256,
.align = 1,
.regset_get = s390_tdb_get,
.set = s390_tdb_set,
},
{
.core_note_type = NT_S390_VXRS_LOW,
.n = __NUM_VXRS_LOW,
.size = sizeof(__u64),
.align = sizeof(__u64),
.regset_get = s390_vxrs_low_get,
.set = s390_vxrs_low_set,
},
{
.core_note_type = NT_S390_VXRS_HIGH,
.n = __NUM_VXRS_HIGH,
.size = sizeof(__vector128),
.align = sizeof(__vector128),
.regset_get = s390_vxrs_high_get,
.set = s390_vxrs_high_set,
[S390] signal race with restarting system calls For a ERESTARTNOHAND/ERESTARTSYS/ERESTARTNOINTR restarting system call do_signal will prepare the restart of the system call with a rewind of the PSW before calling get_signal_to_deliver (where the debugger might take control). For A ERESTART_RESTARTBLOCK restarting system call do_signal will set -EINTR as return code. There are two issues with this approach: 1) strace never sees ERESTARTNOHAND, ERESTARTSYS, ERESTARTNOINTR or ERESTART_RESTARTBLOCK as the rewinding already took place or the return code has been changed to -EINTR 2) if get_signal_to_deliver does not return with a signal to deliver the restart via the repeat of the svc instruction is left in place. This opens a race if another signal is made pending before the system call instruction can be reexecuted. The original system call will be restarted even if the second signal would have ended the system call with -EINTR. These two issues can be solved by dropping the early rewind of the system call before get_signal_to_deliver has been called and by using the TIF_RESTART_SVC magic to do the restart if no signal has to be delivered. The only situation where the system call restart via the repeat of the svc instruction is appropriate is when a SA_RESTART signal is delivered to user space. Unfortunately this breaks inferior calls by the debugger again. The system call number and the length of the system call instruction is lost over the inferior call and user space will see ERESTARTNOHAND/ ERESTARTSYS/ERESTARTNOINTR/ERESTART_RESTARTBLOCK. To correct this a new ptrace interface is added to save/restore the system call number and system call instruction length. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2011-10-30 22:16:47 +08:00
},
{
.core_note_type = NT_S390_HIGH_GPRS,
.n = sizeof(s390_compat_regs_high) / sizeof(compat_long_t),
.size = sizeof(compat_long_t),
.align = sizeof(compat_long_t),
.regset_get = s390_compat_regs_high_get,
.set = s390_compat_regs_high_set,
},
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
{
.core_note_type = NT_S390_GS_CB,
.n = sizeof(struct gs_cb) / sizeof(__u64),
.size = sizeof(__u64),
.align = sizeof(__u64),
.regset_get = s390_gs_cb_get,
s390: add a system call for guarded storage This adds a new system call to enable the use of guarded storage for user space processes. The system call takes two arguments, a command and pointer to a guarded storage control block: s390_guarded_storage(int command, struct gs_cb *gs_cb); The second argument is relevant only for the GS_SET_BC_CB command. The commands in detail: 0 - GS_ENABLE Enable the guarded storage facility for the current task. The initial content of the guarded storage control block will be all zeros. After the enablement the user space code can use load-guarded-storage-controls instruction (LGSC) to load an arbitrary control block. While a task is enabled the kernel will save and restore the current content of the guarded storage registers on context switch. 1 - GS_DISABLE Disables the use of the guarded storage facility for the current task. The kernel will cease to save and restore the content of the guarded storage registers, the task specific content of these registers is lost. 2 - GS_SET_BC_CB Set a broadcast guarded storage control block. This is called per thread and stores a specific guarded storage control block in the task struct of the current task. This control block will be used for the broadcast event GS_BROADCAST. 3 - GS_CLEAR_BC_CB Clears the broadcast guarded storage control block. The guarded- storage control block is removed from the task struct that was established by GS_SET_BC_CB. 4 - GS_BROADCAST Sends a broadcast to all thread siblings of the current task. Every sibling that has established a broadcast guarded storage control block will load this control block and will be enabled for guarded storage. The broadcast guarded storage control block is used up, a second broadcast without a refresh of the stored control block with GS_SET_BC_CB will not have any effect. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
2016-01-26 21:10:34 +08:00
.set = s390_gs_cb_set,
},
{
.core_note_type = NT_S390_GS_BC,
.n = sizeof(struct gs_cb) / sizeof(__u64),
.size = sizeof(__u64),
.align = sizeof(__u64),
.regset_get = s390_gs_bc_get,
.set = s390_gs_bc_set,
},
{
.core_note_type = NT_S390_RI_CB,
.n = sizeof(struct runtime_instr_cb) / sizeof(__u64),
.size = sizeof(__u64),
.align = sizeof(__u64),
.regset_get = s390_runtime_instr_get,
.set = s390_runtime_instr_set,
},
};
static const struct user_regset_view user_s390_compat_view = {
.name = "s390",
.e_machine = EM_S390,
.regsets = s390_compat_regsets,
.n = ARRAY_SIZE(s390_compat_regsets)
};
#endif
const struct user_regset_view *task_user_regset_view(struct task_struct *task)
{
#ifdef CONFIG_COMPAT
if (test_tsk_thread_flag(task, TIF_31BIT))
return &user_s390_compat_view;
#endif
return &user_s390_view;
}
static const char *gpr_names[NUM_GPRS] = {
"r0", "r1", "r2", "r3", "r4", "r5", "r6", "r7",
"r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15",
};
unsigned long regs_get_register(struct pt_regs *regs, unsigned int offset)
{
if (offset >= NUM_GPRS)
return 0;
return regs->gprs[offset];
}
int regs_query_register_offset(const char *name)
{
unsigned long offset;
if (!name || *name != 'r')
return -EINVAL;
if (kstrtoul(name + 1, 10, &offset))
return -EINVAL;
if (offset >= NUM_GPRS)
return -EINVAL;
return offset;
}
const char *regs_query_register_name(unsigned int offset)
{
if (offset >= NUM_GPRS)
return NULL;
return gpr_names[offset];
}
static int regs_within_kernel_stack(struct pt_regs *regs, unsigned long addr)
{
unsigned long ksp = kernel_stack_pointer(regs);
return (addr & ~(THREAD_SIZE - 1)) == (ksp & ~(THREAD_SIZE - 1));
}
/**
* regs_get_kernel_stack_nth() - get Nth entry of the stack
* @regs:pt_regs which contains kernel stack pointer.
* @n:stack entry number.
*
* regs_get_kernel_stack_nth() returns @n th entry of the kernel stack which
* is specifined by @regs. If the @n th entry is NOT in the kernel stack,
* this returns 0.
*/
unsigned long regs_get_kernel_stack_nth(struct pt_regs *regs, unsigned int n)
{
unsigned long addr;
addr = kernel_stack_pointer(regs) + n * sizeof(long);
if (!regs_within_kernel_stack(regs, addr))
return 0;
return *(unsigned long *)addr;
}