Commit 68fc378c authored by Michael Ellerman's avatar Michael Ellerman

Revert "powerpc/tm: Abort syscalls in active transactions"

This reverts commit feba4036.

Although the principle of this change is good, the implementation has a
few issues.

Firstly we can sometimes fail to abort a syscall because r12 may have
been clobbered by C code if we went down the virtual CPU accounting
path, or if syscall tracing was enabled.

Secondly we have decided that it is safer to abort the syscall even
earlier in the syscall entry path, so that we avoid the syscall tracing
path when we are transactional.

So that we have time to thoroughly test those changes we have decided to
revert this for this merge window and will merge the fixed version in
the next window.

NB. Rather than reverting the selftest we just drop tm-syscall from
TEST_PROGS so that it's not run by default.

Fixes: feba4036 ("powerpc/tm: Abort syscalls in active transactions")
Signed-off-by: default avatarMichael Ellerman <mpe@ellerman.id.au>
parent d33047fd
...@@ -74,23 +74,22 @@ Causes of transaction aborts ...@@ -74,23 +74,22 @@ Causes of transaction aborts
Syscalls Syscalls
======== ========
Syscalls made from within an active transaction will not be performed and the Performing syscalls from within transaction is not recommended, and can lead
transaction will be doomed by the kernel with the failure code TM_CAUSE_SYSCALL to unpredictable results.
| TM_CAUSE_PERSISTENT.
Syscalls made from within a suspended transaction are performed as normal and Syscalls do not by design abort transactions, but beware: The kernel code will
the transaction is not explicitly doomed by the kernel. However, what the not be running in transactional state. The effect of syscalls will always
kernel does to perform the syscall may result in the transaction being doomed remain visible, but depending on the call they may abort your transaction as a
by the hardware. The syscall is performed in suspended mode so any side side-effect, read soon-to-be-aborted transactional data that should not remain
effects will be persistent, independent of transaction success or failure. No invisible, etc. If you constantly retry a transaction that constantly aborts
guarantees are provided by the kernel about which syscalls will affect itself by calling a syscall, you'll have a livelock & make no progress.
transaction success.
Care must be taken when relying on syscalls to abort during active transactions Simple syscalls (e.g. sigprocmask()) "could" be OK. Even things like write()
if the calls are made via a library. Libraries may cache values (which may from, say, printf() should be OK as long as the kernel does not access any
give the appearance of success) or perform operations that cause transaction memory that was accessed transactionally.
failure before entering the kernel (which may produce different failure codes).
Examples are glibc's getpid() and lazy symbol resolution. Consider any syscalls that happen to work as debug-only -- not recommended for
production use. Best to queue them up till after the transaction is over.
Signals Signals
...@@ -177,7 +176,8 @@ kernel aborted a transaction: ...@@ -177,7 +176,8 @@ kernel aborted a transaction:
TM_CAUSE_RESCHED Thread was rescheduled. TM_CAUSE_RESCHED Thread was rescheduled.
TM_CAUSE_TLBI Software TLB invalid. TM_CAUSE_TLBI Software TLB invalid.
TM_CAUSE_FAC_UNAV FP/VEC/VSX unavailable trap. TM_CAUSE_FAC_UNAV FP/VEC/VSX unavailable trap.
TM_CAUSE_SYSCALL Syscall from active transaction. TM_CAUSE_SYSCALL Currently unused; future syscalls that must abort
transactions for consistency will use this.
TM_CAUSE_SIGNAL Signal delivered. TM_CAUSE_SIGNAL Signal delivered.
TM_CAUSE_MISC Currently unused. TM_CAUSE_MISC Currently unused.
TM_CAUSE_ALIGNMENT Alignment fault. TM_CAUSE_ALIGNMENT Alignment fault.
......
...@@ -11,7 +11,7 @@ ...@@ -11,7 +11,7 @@
#define TM_CAUSE_RESCHED 0xde #define TM_CAUSE_RESCHED 0xde
#define TM_CAUSE_TLBI 0xdc #define TM_CAUSE_TLBI 0xdc
#define TM_CAUSE_FAC_UNAV 0xda #define TM_CAUSE_FAC_UNAV 0xda
#define TM_CAUSE_SYSCALL 0xd8 #define TM_CAUSE_SYSCALL 0xd8 /* future use */
#define TM_CAUSE_MISC 0xd6 /* future use */ #define TM_CAUSE_MISC 0xd6 /* future use */
#define TM_CAUSE_SIGNAL 0xd4 #define TM_CAUSE_SIGNAL 0xd4
#define TM_CAUSE_ALIGNMENT 0xd2 #define TM_CAUSE_ALIGNMENT 0xd2
......
...@@ -34,7 +34,6 @@ ...@@ -34,7 +34,6 @@
#include <asm/ftrace.h> #include <asm/ftrace.h>
#include <asm/hw_irq.h> #include <asm/hw_irq.h>
#include <asm/context_tracking.h> #include <asm/context_tracking.h>
#include <asm/tm.h>
/* /*
* System calls. * System calls.
...@@ -146,24 +145,6 @@ END_FW_FTR_SECTION_IFSET(FW_FEATURE_SPLPAR) ...@@ -146,24 +145,6 @@ END_FW_FTR_SECTION_IFSET(FW_FEATURE_SPLPAR)
andi. r11,r10,_TIF_SYSCALL_DOTRACE andi. r11,r10,_TIF_SYSCALL_DOTRACE
bne syscall_dotrace bne syscall_dotrace
.Lsyscall_dotrace_cont: .Lsyscall_dotrace_cont:
#ifdef CONFIG_PPC_TRANSACTIONAL_MEM
BEGIN_FTR_SECTION
b 1f
END_FTR_SECTION_IFCLR(CPU_FTR_TM)
extrdi. r11, r12, 1, (63-MSR_TS_T_LG) /* transaction active? */
beq+ 1f
/* Doom the transaction and don't perform the syscall: */
mfmsr r11
li r12, 1
rldimi r11, r12, MSR_TM_LG, 63-MSR_TM_LG
mtmsrd r11, 0
li r11, (TM_CAUSE_SYSCALL|TM_CAUSE_PERSISTENT)
TABORT(R11)
b .Lsyscall_exit
1:
#endif
cmpldi 0,r0,NR_syscalls cmpldi 0,r0,NR_syscalls
bge- syscall_enosys bge- syscall_enosys
......
TEST_PROGS := tm-resched-dscr tm-syscall TEST_PROGS := tm-resched-dscr
all: $(TEST_PROGS) all: $(TEST_PROGS)
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment