From: Julian Seward Date: Thu, 13 Aug 2015 12:55:09 +0000 (+0000) Subject: Update following amd64 AVX front end fixes. X-Git-Tag: svn/VALGRIND_3_11_0~124 X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=a34651d40c15b357e9f477748496dc89875ef59d;p=thirdparty%2Fvalgrind.git Update following amd64 AVX front end fixes. git-svn-id: svn://svn.valgrind.org/valgrind/trunk@15533 --- diff --git a/NEWS b/NEWS index b0e2a12cbf..69bfdfae3c 100644 --- a/NEWS +++ b/NEWS @@ -139,7 +139,9 @@ where XXXXXX is the bug number as listed below. == 307399 == 343175 == 342740 + == 346912 335907 segfault when running wine's ddrawex/tests/surface.c under valgrind +338602 AVX2 bit in CPUID missing 338606 Strange message for scripts with invalid interpreter 338731 ppc: Fix testuite build for toolchains not supporting -maltivec 338995 shmat with hugepages (SHM_HUGETLB) fails with EINVAL @@ -172,6 +174,10 @@ where XXXXXX is the bug number as listed below. 342063 wrong format specifier for test mcblocklistsearch in gdbserver_tests 342221 socket connect false positive uninit memory for unknown af family 342353 Allow dumping full massif output while valgrind is still running +342571 Valgrind chokes on AVX compare intrinsic with _CMP_GE_QS + == 346476 + == 348387 + == 350593 342603 Add I2C_SMBUS ioctl support 342635 OS X 10.10 (Yosemite) - missing system calls and fcntl code 342683 Mark memory past the initial brk limit as unaddressable diff --git a/docs/internals/3_10_BUGSTATUS.txt b/docs/internals/3_10_BUGSTATUS.txt index 90f4ba1786..a9d1870895 100644 --- a/docs/internals/3_10_BUGSTATUS.txt +++ b/docs/internals/3_10_BUGSTATUS.txt @@ -4,10 +4,6 @@ go in here. === VEX/amd64 ========================================================== -338602 AVX2 bit in CPUID missing - need to do XSAVE, XRSTOR - == 338602 - 339596 vex amd64->IR: 0x8F 0xE8 0x78 0xCD 0xC1 0x4 0xC5 0xF9 XOP/FMA -- should take @@ -18,30 +14,15 @@ go in here. 340469 vex amd64->IR: unhandled instruction bytes: 0x66 0xA7 Not clear what's going on here -342571 Valgrind chokes on AVX compare intrinsic with _CMP_GE_QS - probably easy to fix - 342988 INT 80 not recognised on amd64 -346476 vex amd64->IR: 0xC5 0xFD 0xC2 0xC1 0x10 0xC5 0xFD 0x29 - (various AVX CMP variants) - -346912 callgrind bails out on vpmaskmovd instruction - I think this is fixed now. Need to check. - -347698 3.10.1 doesn't recognize cmpsd generated by DynASM - 347885 unhandled instruction bytes: 0xD6 0xF 0x84 0x1B (SALC amd64) -348387 Valgrind does not recognize a variant of the VCMPPD instruction - 348574 vex amd64->IR pcmpistri SSE4.2 instruction is unsupported 349304 vex amd64->IR: 0xF 0x33 0x89 0x45 0xFC 0x89 0x55 0xF8 rdpmc? -350593 AMD64 VEX does not support all SIMD compare variants. - === VEX/arm ============================================================ 335618 mov.w rN, pc/sp (ARM32)